1. Introduction
The growing demand for lightweight commercial vehicles has driven continuous innovation in structural component manufacturing. In my research, I focused on the integrated design and casting process optimization of a ductile iron crossbeam used in commercial vehicle chassis. Traditional steel-welded crossbeams typically involve multiple stamping and assembly processes, leading to high production costs, ecreased structural integrity, and increased vehicle mass. As an alternative, ductile iron offers excellent castability, high strength-to-weight efficiency, and outstanding fatigue resistance, making it an ideal candidate for integrated structural components.
My work aims to develop a comprehensive methodology that combines structural integrated design, numerical simulation of the casting process, defect prediction, and experimental verification. The primary goal was to minimize casting defects, including shrinkage cavities, shrinkage porosity, and gas entrapment, while achieving a 28.7 % weight reduction compared with the original fabricated structure. This paper systematically details my research procedure, including structural static analysis, casting process design, orthogonal experiments, and material characterization.
2. Materials and Experimental Procedures
2.1 Raw Materials and Alloy Design
In my experiments, I selected QT800-5 ductile iron as the target material due to its excellent combination of strength and ductility. The raw materials consisted of high-purity Q10 pig iron (low sulfur, low manganese), carbon steel scrap, return scrap, FeSiMg6RE2 rare-earth magnesium nodulizer, and a Si-Ba inoculant. The chemical composition of the main raw materials is presented in Table 1.
Table 1 Chemical Composition of Raw Materials (wt.%)
| Material | C | Si | Mn | P | S | RE | Mg | Ba | Fe |
|—|—|—|—|—|—|—|—|—|—|
| Q10 Pig iron | 4.25 | 0.23 | 0.05 | 0.018 | 0.010 | – | – | – | Balance |
| Steel scrap | 0.26 | 0.22 | 0.20 | 0.020 | 0.010 | – | – | – | Balance |
| Nodulizer | – | 41–44 | 2–3 | – | – | 6–7 | – | – | Balance |
| Si-Ba inoculant | – | 60–65 | – | – | – | – | – | 4–6 | Balance |
| 75SiFe | – | 72–80 | – | – | – | – | – | – | Balance |
The target chemical composition was carefully controlled to ensure stable nodularization and adequate pearlite formation. The base iron was melted in a 1000 kg medium-frequency induction furnace. I controlled the tapping temperature within the range of 1490–1520 °C to minimize oxidation and nodulizer fading. The nodularization treatment was performed using the cored wire injection method, achieving a magnesium recovery rate of approximately 50%.
2.2 Melting and Inoculation Process
The melting procedure I adopted involved the following steps:
(1) Charge calculation: In order to achieve the target composition for QT800-5, I calculated the weight ratio of pig iron, scrap steel, recarburizer, copper, nickel, and ferromolybdenum.
(2) Melting: High-purity pig iron, steel scrap, and recarburizer were charged into the furnace and melted rapidly to prevent excessive oxidation. When the melt temperature reached 1530–1570 °C, I conducted slagging and high-temperature holding to purify the melt. Subsequently, electrolytic copper and nickel plates were added, and the melt temperature was adjusted to 1490–1520 °C before tapping.
(3) Nodularization and inoculation: A three-stage inoculation scheme was used to maximize the nodule count and minimize casting defects. First, Si-Ba inoculant (0.3–0.4 wt.%) was placed at the bottom of the pouring ladle for initial inoculation. Second, during the transfer of two-thirds of the molten iron, an additional 0.2–0.3 wt.% inoculant was added. Finally, stream inoculation was performed during pouring using 75SiFe (0.1–0.2 wt.%).
(4) Pouring: The molten iron was poured at a temperature between 1380 °C and 1420 °C into resin sand molds. The pouring was completed within 10 minutes after nodularization to avoid the fading of the nodulizing and inoculating effects.
2.3 Finite Element Method Fundamentals
In my numerical simulation, I applied the finite element method (FEM) to solve the partial differential equations governing fluid flow, heat transfer, and solidification. The general governing equation for the transient heat conduction problem can be written as:
$$\rho c_p \frac{\partial T}{\partial t} =
abla \cdot (k
abla T) + Q$$
where $\rho$ is the density, $c_p$ is the specific heat capacity at constant pressure, $T$ is the temperature, $t$ is the time, $k$ is the thermal conductivity, and $Q$ is the latent heat source term. For the flow field calculation, I employed the Navier-Stokes equations combined with the volume-of-fluid (VOF) method to track the free surface during mold filling:
$$\rho\left(\frac{\partial \mathbf{u}}{\partial t} + \mathbf{u} \cdot
abla \mathbf{u}\right) = –
abla p + \mu
abla^2 \mathbf{u} + \rho \mathbf{g}$$
where $\mathbf{u}$ is the velocity vector, $p$ is the pressure, $\mu$ is the dynamic viscosity, and $\mathbf{g}$ is the gravitational acceleration.
2.4 Characterization Methods
To verify the casting quality, I systematically characterized the chemical composition, microstructure, and mechanical properties of the integrated crossbeam castings. Optical emission spectroscopy (OES) was used to measure the chemical composition. Metallographic specimens were prepared by standard grinding and polishing procedures, followed by etching with 4% nital solution. I used a Zeiss Axiolab5 optical microscope and Image-Pro Plus image analysis software for quantitative metallography. The graphite nodularity and nodule size were evaluated according to Chinese national standards GB/T 9941-2021.
For mechanical property evaluation, I performed tensile tests using a SHIMADZU AGS-X universal testing machine, and hardness measurements using an OVH-5 Vickers hardness tester. Fracture surface analysis was conducted using a GeminiSEM 300 field emission scanning electron microscope equipped with energy-dispersive X-ray spectroscopy (EDS).
3. Structural Integration and Static Analysis of the Crossbeam
3.1 Original Crossbeam Structure
The original crossbeam structure in my study weighed 50.5 kg and consisted of seven components: four L-shaped steel plates, two mounting brackets (QT550 ductile iron), and one U-shaped steel plate fabricated from B510L steel. The key technical parameters are given in Table 2.
Table 2 Technical Parameters of the Original Crossbeam
| Component | Material | Elastic modulus (MPa) | Yield strength (MPa) | Poisson’s ratio | Mass (kg) | Overall dimensions (mm) |
|—|—|—|—|—|—|—|
| Mounting bracket (×2) | QT550 | 172000 | 320 | 0.275 | 9.78 × 2 | 814×237×600 |
| Steel plates | B510L | 200000 | 355 | 0.3 | (4.46+5.07)×2 | – |
This multi-piece design requires extensive stamping, machining, and welding operations, which are both labor-intensive and inefficient. Moreover, stress concentration at welded joints and bolt holes significantly compromises structural reliability.
3.2 Static Analysis of the Original Design
To establish a performance baseline, I performed static finite element analysis on the original crossbeam structure using Hypermesh software. The finite element model incorporated shell elements for the frame rails and crossbeam plates, and solid tetrahedral elements for the cast brackets. The model consisted of 85122 elements and 87864 nodes for the frame, and 96840 elements and 24638 nodes for the crossbeam castings. Bolt connections were simulated using RBE2 rigid elements.
Static analysis was performed under four critical working conditions:
(1) Acceleration condition: A driving force of 60000 N was applied along the thrust rod (-x direction), with fully constrained wheel mounting nodes. The maximum equivalent stress was calculated as 85.1 MPa, located at the threaded hole region.
(2) Turning condition: A lateral force of 80000 N was applied in the -y direction. The maximum equivalent stress was 396.1 MPa, also at the threaded hole area.
(3) Torsion condition: The maximum torque condition of 80000 N·m was simulated as equivalent equal-and-opposite forces at both ends of the crossbeam. The maximum equivalent stress was 680.6 MPa, exceeding the yield strength of B510L steel (355 MPa), which poses a significant durability risk.
(4) Emergency braking condition: Equivalent forces of 20000 N in opposite directions were applied. The maximum equivalent stress was 95.4 MPa.
The static analysis results revealed that the threaded hole regions are the most vulnerable areas, subject to severe stress concentration, especially under torsional loading.
3.3 Integrated Structure Design
Based on the static analysis results and structural design principles, I redesigned the original multi-component crossbeam into a single integrated ductile iron casting. The design principles I followed included:
(a) Minimum wall thickness principle: The casting wall thickness should be as thin as possible while maintaining adequate strength and rigidity, thereby reducing weight and improving filling capability.
(b) Rib design flexibility: The rib structure should be designed to avoid stress concentration and provide adequate stiffness without creating excessively thick sections.
(c) Stress-oriented design: The structural geometry should be optimized to achieve uniform stress distribution and eliminate sharp corners and discontinuities.
Using the equal-stiffness substitution method, I calculated the minimum wall thickness for ductile iron as follows:
$$E_{sl} \cdot t_{sl}^3 = E_{dl} \cdot t_{dl}^3$$
where $E_{sl}$ and $E_{dl}$ are the elastic moduli of steel and ductile iron, respectively, and $t_{sl}$ and $t_{dl}$ are the corresponding wall thicknesses. The calculated minimum wall thickness for ductile iron was 8.0 mm.
The integrated crossbeam design features a circular-axis cross-section to enhance torsional rigidity, replacing the original I-shaped steel plate structure. The resulting integrated model had overall dimensions of 839 × 403 × 221 mm and a mass of 36 kg, achieving a weight reduction of 28.7% compared with the original 50.5 kg structure.
3.4 Static Analysis of the Integrated Crossbeam
The integrated crossbeam was placed within the original frame environment for static analysis. The finite element model comprised 66444 shell elements and 67864 nodes for the frame, and 3309713 tetrahedral elements and 133147 nodes for the crossbeam casting. The same load cases and boundary conditions were applied.
Table 3 Comparison of Maximum Equivalent Stresses between Original and Integrated Designs
| Working condition | Original max. stress (MPa) | Integrated max. stress (MPa) |
|—|—|—|
| Acceleration | 85.1 | 69.3 |
| Turning | 396.1 | 311.1 |
| Torsion | 680.6 | 558.4 |
| Emergency braking | 95.4 | 95.4 |
The maximum equivalent stress observed in the integrated design was 558.4 MPa under torsion, located at the threaded hole area. With a material yield strength of at least 800 MPa for QT800-5, the minimum safety factor was calculated as:
$$n = \frac{\sigma_y}{\sigma_{\text{max}}} = \frac{800 \text{ MPa}}{558.4 \text{ MPa}} = 1.43$$
This safety factor of 1.4 exceeds the typical requirement of 1.4 for commercial vehicle components, validating the structural adequacy of my integrated design.
4. Casting Process Design and Numerical Simulation
4.1 Parting Line Selection
Based on the structural characteristics of the integrated crossbeam, I chose the maximum cross-section contour as the parting line. This selection ensures easy mold release, simplified mold structure, and acceptable dimensional accuracy (CT10 grade).
4.2 Sand Core Design
The complex internal cavity of the crossbeam required a resin-bonded sand core. I designed a unified sand core with a venting system to facilitate the removal of gases generated during pouring. The core was designed with dimensions corresponding to the internal geometry of the casting, ensuring proper cavity formation and dimensional accuracy.
4.3 Gating System Design
I designed a closed gating system with a cross-sectional area ratio of F_direct:F_running:F_inner = 2.50:1.25:1. Four inner gates were arranged symmetrically to ensure uniform filling of the large-span crossbeam. The cross-sectional area of the inner gate was calculated using the flow rate method:
$$A_g = \frac{V}{v_g \times t}$$
where $V$ is the volume of molten metal passing through the gate (mm³), $v_g$ is the flow velocity (m/s), and $t$ is the filling time (s). The filling time was calculated using the empirical formula:
$$t = f\left(\sqrt{G} + \sqrt{0.2WG^3}\right)$$
For $f = 0.7$ (material coefficient), $G = 36$ kg (casting mass), and $W = 8.0$ mm (wall thickness), the calculated filling time was 7.27 s, corresponding to a pouring speed of 11 cm/s.
4.4 Initial Simulation Results
The initial casting process was simulated using a commercial finite element-based simulation software package. The mesh size was set to 2 mm, resulting in a total of 12,266,795 cells. The physical properties of QT800-5 used in the simulation are listed in Table 4.
Table 4 Physical Properties of QT800-5 Used in Simulation
| Property | Value |
|—|—|
| Density (kg/m³) | 6459 |
| Liquidus temperature (K) | 1457 |
| Solidus temperature (K) | 1418 |
| Latent heat (kJ/kg) | 210 |
| Specific heat capacity (kJ/kg·K) | 0.78 |
| Heat transfer coefficient (W/m²·K) | 700 |
The initial simulation was performed with a pouring temperature of 1400 °C, a pouring speed of 11 cm/s, and a mold temperature of 20 °C. The filling process analysis showed that the molten iron filled the cavity steadily, but the solidification analysis revealed isolated liquid pools at both ends of the casting. These regions exhibited the longest solidification times and were identified as potential locations for shrinkage cavities and porosity.
The initial design produced total shrinkage porosity of 751.16 mm³. The liquid-phase distribution confirmed the formation of closed liquid regions at the thick-walled ends of the crossbeam, where melt replenishment was restricted. These regions represent a high risk of casting defects, including shrinkage cavities and shrinkage porosity.
4.5 Optimization of the Gating System
To improve the casting quality and reduce casting defects, I optimized the riser system by increasing the number of risers from 4 to 8 (4 × Φ70 mm × 120 mm) and rationally distributing them at both ends of the crossbeam. The simulation results for the optimized system showed:
Table 5 Comparison of Initial and Optimized Gating Systems
| Parameter | Initial design | Optimized design |
|—|—|—|
| Number of risers | 4 | 8 (4 × Φ70 mm × 120 mm) |
| Solidification time (s) | 436 | 307.67 |
| Isolated liquid regions | Present | Eliminated |
| Shrinkage porosity volume (mm³) | 751.16 | 685.85 |
| Defect reduction (%) | – | 8.69 |
In the optimized system, the last solidifying regions were confined to the risers and gating system components, confirming that proper feeding was established. The shrinkage porosity defects were largely relocated to the risers, which fulfills their intended function of collecting shrinkage porosity and oxides, thereby reducing casting defects in the final product.
5. Numerical Simulation of Filling and Solidification under Different Process Parameters
5.1 Effect of Pouring Temperature
To investigate the effect of pouring temperature on filling behavior and solidification characteristics, I conducted simulations at 1380 °C, 1400 °C, and 1420 °C, while maintaining the pouring speed at 11 cm/s and mold temperature at 20 °C.
5.1.1 Filling Simulation Results
The filling simulation results for the three pouring temperatures are summarized in Table 6. At lower pouring temperatures, the fluidity of the molten iron decreased, resulting in slower filling of the complex sections. Conversely, at 1420 °C, the melt exhibited enhanced fluidity, yet the flow pattern became more turbulent, increasing the likelihood of gas entrapment.
Table 6 Filling Simulation Results at Different Pouring Temperatures
| Pouring temperature (°C) | Pouring time (s) | Minimum temperature after filling (°C) | Air entrapment volume (mm³) |
|—|—|—|—|
| 1380 | 7.27 | 1258 | 1700.58 |
| 1400 | 7.27 | 1275 | 1682.44 |
| 1420 | 7.27 | 1288 | 1697.82 |
The air entrapment volume followed a nonlinear trend as the pouring temperature increased. When the temperature increased from 1380 °C to 1400 °C, the air entrapment decreased because the improved fluidity promoted smooth filling and reduced turbulence-induced gas entrapment. However, when the temperature was further increased to 1420 °C, the air entrapment increased because the surface tension of the molten iron decreased, and gas bubbles were more easily dispersed and entrapped in the melt.
5.1.2 Solidification Simulation Results
The solidification simulation results at different pouring temperatures are presented in Table 7. The total solidification time increased with increasing pouring temperature, and the shrinkage porosity volume showed a monotonically increasing trend.
Table 7 Solidification Simulation Results at Different Pouring Temperatures
| Pouring temperature (°C) | Total solidification time (s) | Shrinkage porosity volume (mm³) |
|—|—|—|
| 1380 | 307.67 | 667.92 |
| 1400 | 326.96 | 685.85 |
| 1420 | 346.33 | 702.11 |
The increase in shrinkage porosity volume with increasing pouring temperature is attributed to higher thermal contraction and wider solidification temperature ranges, which adversely affect the feeding capability of the risers. These results highlight that for high-quality ductile iron castings, pouring temperature must be carefully balanced to minimize casting defects.
5.2 Effect of Pouring Speed
I also studied the effect of pouring speed on the filling behavior and solidification characteristics by varying the pouring speed from 10 to 12 cm/s, while maintaining the pouring temperature at 1400 °C and the mold temperature at 20 °C.
5.2.1 Filling Simulation Results
The filling simulation results for different pouring speeds are shown in Table 8.
Table 8 Filling Simulation Results at Different Pouring Speeds
| Pouring speed (cm/s) | Filling time (s) | Air entrapment volume (mm³) |
|—|—|—|
| 10 | 8.00 | 1655.77 |
| 11 | 7.27 | 1682.44 |
| 12 | 6.36 | 1700.39 |
There was a clear positive correlation between pouring speed and air entrapment. At higher pouring speed, the kinetic energy of the molten iron increased, leading to enhanced shear forces at the cavity walls, boundary layer separation, vortex formation, and three-dimensional flow patterns. These flow instabilities entrapped air into the bulk melt, resulting in higher air entrapment volumes.
5.2.2 Solidification Simulation Results
The solidification simulation results at different pouring speeds are shown in Table 9.
Table 9 Solidification Simulation Results at Different Pouring Speeds
| Pouring speed (cm/s) | Total solidification time (s) | Shrinkage porosity volume (mm³) |
|—|—|—|
| 10 | 305.69 | 696.92 |
| 11 | 326.96 | 685.85 |
| 12 | 348.38 | 680.85 |
The shrinkage porosity volume decreased as the pouring speed increased. This is because faster filling resulted in more uniform temperature distribution within the mold and reduced heat loss during filling, leading to better riser feeding efficiency. Conversely, lower filling speeds increased the cooling rate of the melt, resulting in insufficient feeding and more shrinkage porosity.
6. Orthogonal Experiment for Process Parameter Optimization
6.1 Orthogonal Array Design
To systematically optimize the casting process parameters and minimize casting defects, I adopted a two-factor, three-level orthogonal experimental design. The factors and levels are listed in Table 10.
Table 10 Factor-Level Table for Orthogonal Experiment
| Factor | Level 1 | Level 2 | Level 3 |
|—|—|—|—|
| Pouring speed A (cm/s) | 10 | 11 | 12 |
| Pouring temperature B (°C) | 1380 | 1400 | 1420 |
The complete orthogonal experiment design is shown in Table 11, including the response variables of air entrapment volume and shrinkage porosity volume.
Table 11 Orthogonal Experiment Design and Results
| Experiment No. | Pouring speed (cm/s) | Pouring temperature (°C) | Air entrapment (mm³) | Shrinkage porosity (mm³) |
|—|—|—|—|—|
| Y1 | 10 | 1380 | 1677.19 | 679.18 |
| Y2 | 10 | 1400 | 1655.77 | 696.92 |
| Y3 | 10 | 1420 | 1668.84 | 713.20 |
| Y4 | 11 | 1400 | 1682.44 | 685.85 |
| Y5 | 11 | 1420 | 1697.82 | 702.11 |
| Y6 | 11 | 1380 | 1700.58 | 667.92 |
| Y7 | 12 | 1420 | 1704.32 | 697.20 |
| Y8 | 12 | 1400 | 1700.39 | 680.85 |
| Y9 | 12 | 1380 | 1721.51 | 663.00 |
6.2 Range Analysis for Air Entrapment
The range analysis results for air entrapment volume are summarized in Table 12.
Table 12 Range Analysis Results for Air Entrapment Volume
| Factor | A (pouring speed) | B (pouring temperature) |
|—|—|—|
| K1 (Level 1 sum) | 5001.80 | 5099.28 |
| K2 (Level 2 sum) | 5080.84 | 5038.60 |
| K3 (Level 3 sum) | 5126.22 | 5070.98 |
| Range R | 124.42 | 60.68 |
| Rank | 1 | 2 |
The range value for pouring speed (124.42) was significantly larger than that for pouring temperature (60.68), indicating that pouring speed has a more pronounced effect on air entrapment than pouring temperature.
6.3 Statistical Analysis of Variance
To rigorously evaluate the significance of each factor, I performed one-way analysis of variance (ANOVA). The F-value and p-value were calculated using the standard ANOVA procedure:
$$F = \frac{MS_{\text{factor}}}{MS_{\text{error}}}$$
where $MS_{factor}$ is the mean square for the factor and $MS_{error}$ is the error mean square.
Table 13 ANOVA Results for Air Entrapment
| Source | Sum of squares | DF | Mean square | F-value | p-value |
|—|—|—|—|—|—|
| Pouring speed | 2643.000 | 2 | 1321.500 | 24.266 | 0.006 |
| Pouring temperature | 567.608 | 2 | 283.804 | 5.211 | 0.077 |
| Error | 108.919 | 2 | 54.459 | – | – |
The p-value for pouring speed was 0.006 (< 0.05), indicating a statistically significant effect on air entrapment, while pour temperature showed no significant effect in this test range.
Table 14 ANOVA Results for Shrinkage Porosity
| Source | Sum of squares | DF | Mean square | F-value | p-value |
|—|—|—|—|—|—|
| Pouring speed | 406.121 | 2 | 203.061 | 2.847 | 0.167 |
| Pouring temperature | 1747.142 | 2 | 873.571 | 12.247 | 0.017 |
| Error | 142.659 | 2 | 71.330 | – | – |
For shrinkage porosity, the pouring temperature showed a significant effect (p = 0.017 < 0.05), while pouring speed was not statistically significant. Therefore, the critical process parameters for controlling casting defects are different for different defect types.
6.4 Determination of Optimal Parameters
Based on the range analysis and ANOVA results, I established the following response trends:
For air entrapment, the optimal combination is A1 (10 cm/s) and B2 (1400 °C), which minimizes gas entrapment defects.
For shrinkage porosity, the optimal combination is A3 (12 cm/s) and B1 (1380 °C), which minimizes the shrinkage porosity volume.
However, considering the comprehensive quality requirements, I selected the following compromise: pouring speed of 11 cm/s, pouring temperature of 1380 °C, and mold temperature of 20 °C. This combination achieved the lowest shrinkage porosity (667.92 mm³) with acceptable air entrapment, effectively balancing the two types of casting defects.
7. Production Trial and Performance Characterization
7.1 Experimental Validation
Based on the optimized casting process parameters, I conducted production trials of the integrated ductile iron crossbeam. The optimized gating system is illustrated in the experimental setup. The molten iron was prepared according to the previously described melting and nodularization procedures, and cast at 1380 °C with a pouring speed of 11 cm/s. After solidification and cooling, the castings were subjected to sand removal, gate cutting, shot blasting, and surface finishing.
7.2 Chemical Composition Analysis
The chemical composition of specimens taken from three different positions of the crossbeam was analyzed using emission spectrometry, and the results are shown in Table 15.
Table 15 Chemical Composition of Integrated Crossbeam Specimens
| Element | C | Si | Mn | P | S | Mg | Ni | Mo | Cu |
|—|—|—|—|—|—|—|—|—|—|
| Content (wt.%) | 3.61 | 2.67 | 0.105 | 0.007 | 0.007 | 0.043 | 0.438 | 0.151 | 0.803 |
The composition was stable, with sufficiently low phosphorus (0.007%) and sulfur (0.007%) contents, thereby minimizing the risk of phosphide eutectics and sulfide inclusions. The carbon equivalent was calculated as:
$$\text{CE} = \text{C} + \frac{1}{3}(\text{Si} + \text{P}) = 3.61 + \frac{1}{3}(2.67 + 0.007) = 4.50\%$$
This carbon equivalent value falls within the typical range for ductile iron, promoting good graphitization and reducing the tendency for shrinkage porosity.
7.3 Microstructural Analysis
Microstructural specimens were prepared from three representative locations: position 1 (near gate), position 2 (thin-walled shaft), and position 3 (far gate). The metallographic analysis results are presented in Table 16.
Table 16 Metallographic Rating Results of Crossbeam Specimens
| Specimen | Nodularity grade | Graphite size grade | Pearlite content (%) | Carbide (%) | Phosphide (%) |
|—|—|—|—|—|—|
| Position 1 | 2 | 6 | ≥90 | 1 | 0 |
| Position 2 | 2 | 7 | ≥95 | 1 | 0 |
| Position 3 | 2 | 6 | ≥90 | 1 | 0 |
All specimens exhibited a nodularity grade of 2 (nodularity ≥ 90%) and graphite sizes ranging from grade 6–7. The graphite at position 1 (near gate, slower cooling) exhibited the largest average diameter and the lowest nodule count of 180 nodules/mm². Position 2 (thin-walled region, fastest cooling) exhibited the smallest graphite size and the highest nodule count of 252 nodules/mm². The pearlite content was consistently above 90% in all specimens, confirming that the desired high-strength matrix structure was achieved.
X-ray radiographic inspection of the far-gate regions and the thin-walled central region revealed no cracks, gas pores, or shrinkage defects, confirming the soundness of the critical structural areas.
7.4 Mechanical Performance
The tensile properties of specimens from the three positions are summarized in Table 17.
Table 17 Tensile Test Results of Integrated Crossbeam at Different Positions
| Specimen | Yield strength Rp0.2 (MPa) | Ultimate tensile strength Rm (MPa) | Elongation A (%) |
|—|—|—|—|
| Position 1 | 546 | 875 | 7.6 |
| Position 2 | 562 | 891 | 6.0 |
| Position 3 | 552 | 854 | 6.5 |
The position 1 specimen exhibited the lowest yield strength (546 MPa) but the highest elongation (7.6%) due to slower cooling and coarser graphite. The position 2 specimen exhibited the highest strength values (yield strength of 562 MPa, ultimate tensile strength of 891 MPa) but the lowest elongation (6.0%). The position 3 specimen showed intermediate properties with an ultimate tensile strength of 854 MPa.
The Vickers hardness values of the three specimens are shown in Table 18.
Table 18 Hardness Measurements at Different Positions (HV)
| Specimen | Measurement 1 | Measurement 2 | Measurement 3 | Average |
|—|—|—|—|—|
| Position 1 | 280.8 | 285.8 | 285.8 | 284.1 |
| Position 2 | 272.2 | 293.8 | 279.7 | 281.9 |
| Position 3 | 297.3 | 280.8 | 279.7 | 285.9 |
All hardness values exceeded 270 HV, satisfying the application requirements. The small coefficient of variation among the three positions indicated uniform microstructural properties throughout the casting.
Table 19 Comparison of Measured Properties with QT800-5 Requirements
| Property | QT800-5 Requirement | Measured Minimum | Verification |
|—|—|—|—|
| Ultimate tensile strength (MPa) | ≥ 800 | 854 | ✓ Satisfied |
| Yield strength (MPa) | ≥ 500 | 546 | ✓ Satisfied |
| Elongation (%) | ≥ 5 | 6.0 | ✓ Satisfied |
| Hardness (HV) | ≥ 270 | 281.9 | ✓ Satisfied |
The measured properties for all positions met the QT800-5 requirements, with a safety margin that ensures the reliable performance of the integrated crossbeam in actual service.
7.5 Fracture Morphology and EDS Analysis
The fracture surfaces of tensile specimens were examined by SEM. The fracture exhibited a typical quasi-cleavage mechanism characterized by cleavage facets and river patterns. The size of the cleavage facets and the extent of the river patterns increased from position 1 to position 3, indicating reduced ductility at the far-gate location.
EDS analysis identified the presence of calcium oxide inclusions, which acted as crack initiation sites and significantly diminished the mechanical properties of the casting. Notably, these angular oxide inclusions were more detrimental to the mechanical performance than the spherical graphite particles, demonstrating the importance of melt cleanliness for minimizing casting defects.
8. Discussion
8.1 Relationship Between Process Parameters and Casting Defects
My study demonstrated that the formation of casting defects in the integrated ductile iron crossbeam is primarily governed by the interplay between filling kinetics and solidification characteristics.
Table 20 Correlation Matrix between Process Parameters and Casting Defects
| Parameter | Relationship with air entrapment | Relationship with shrinkage porosity | Optimal level |
|—|—|—|—|
| Pouring speed | Positive correlation (higher speed, more entrapment) | Negative correlation (higher speed, less porosity) | 11 cm/s (compromise) |
| Pouring temperature | Valley-shaped relationship (minimum at 1400 °C) | Positive correlation (higher temperature, more porosity) | 1380 °C (for porosity); 1400 °C (for air entrapment) |
The positive correlation between pouring speed and air entrapment arises from the increased kinetic energy of the melt, which promotes turbulent flow and vortex formation. When the filling speed is excessive, the melt front becomes unstable, leading to the folding and entrapment of mold gases. Conversely, a higher filling speed reduces heat losses and creates a more uniform temperature field, facilitating riser feeding and reducing shrinkage porosity.
The valley-shaped relationship between pouring temperature and air entrapment has a dual origin. On one hand, higher temperature reduces viscosity and improves filling fluidity, which decreases turbulence-induced gas entrapment. On the other hand, excessively high temperature reduces surface tension and increases the vapor pressure of alloying elements, promoting gas nucleation and entrapment.
8.2 Solidification Characteristics and Feeding Strategy
The solidification simulation results showed that the shrinkage porosity volume positively correlates with pouring temperature. This relationship can be quantified using the volumetric shrinkage model:
$$\Delta V_s = V_s \cdot \varepsilon_s = V_s \left( \alpha_l \Delta T + \varepsilon_f \right)$$
where $\Delta V_s$ is the volumetric shrinkage, $\varepsilon_s$ is the linear shrinkage coefficient, $\alpha_l$ is the liquid thermal expansion coefficient, $\Delta T$ is the superheat, and $\varepsilon_f$ is the solidification contraction. Higher pouring temperatures increase the superheat term, resulting in greater total shrinkage and increased porosity.
Ductile iron exhibits mushy solidification characteristics because its co-eutectic solidification occurs over a wider temperature range than gray iron. The graphite nodules nucleate and grow within a continuously changing liquid matrix, encapsulated by an austenite shell. The carbon diffusion through this shell limits the graphite growth rate, and the solidification is completed by continuous nucleation of new grains. This mushy solidification mode makes ductile iron particularly prone to forming isolated liquid pools at final solidification stages. I addressed this by appropriately locating risers and controlling the pouring speed so that these isolated regions are replenished by liquid iron before complete solidification.
8.3 Role of Inoculation and Nodularization
The metallographic results revealed
uniformly distributed spheroidal graphite without significant carbide formation, attesting to an effective inoculation scheme. The magnesium recovery of 0.043 wt.% was sufficient to achieve complete nodularization while being low enough to avoid excessive carbide stabilization. The pearlite content of over 90% in all positions was achieved through the addition of alloying elements, particularly copper and nickel, which promote a pearlitic matrix by retarding ferrite nucleation.
9. Conclusions
In this comprehensive investigation, I developed a complete methodology for the integrated design and casting process optimization of a ductile iron commercial vehicle crossbeam. The main conclusions are as follows:
(1) Structural integration: The original seven-component steel-welded crossbeam was redesigned as a single integrated ductile iron casting, achieving a weight reduction from 50.5 kg to 36 kg (28.7 % decrease). Static analysis under four critical working conditions showed a maximum equivalent stress of 558.4 MPa, which corresponds to a safety factor of 1.43 for QT800-5 material, satisfying commercial vehicle strength requirements.
(2) Gating system optimization: By increasing the number of risers from 4 to 8 and strategically positioning them at both ends of the crossbeam, I successfully eliminated isolated liquid pools and improved the feeding efficiency during solidification. The optimized riser system reduced shrinkage porosity from 751.16 mm³ to 685.85 mm³, a reduction of 8.69%.
(3) Process parameter elucidation: Through systematic numerical simulation, I established the following quantitative relationships:
– Air entrapment volume increased monotonically with pouring speed (from 1655.77 mm³ at 10 cm/s to 1700.39 mm³ at 12 cm/s).
– Air entrapment showed a valley-shaped dependence on pouring temperature, reaching a minimum of 1682.44 mm³ at 1400 °C.
– Shrinkage porosity decreased with increasing pouring speed (from 696.92 to 680.85 mm³) but increased with increasing pouring temperature (from 667.92 to 702.11 mm³).
(4) Statistical optimization: The variance analysis demonstrated that pouring speed significantly affects air entrapment (p = 0.006), while pouring temperature significantly affects shrinkage porosity (p = 0.017). The optimized process parameters are pouring speed of 11 cm/s, pouring temperature of 1380 °C, and mold temperature of 20 °C, which effectively minimize casting defects.
(5) Experimental validation: Production trials confirmed the reliability of the optimized casting process. The castings exhibited stable chemical composition, uniform microstructure, and excellent mechanical properties. All three representative positions within the casting met the QT800-5 requirements, with ultimate tensile strength ≥ 854 MPa, yield strength ≥ 546 MPa, elongation ≥ 6.0%, and hardness ≥ 270 HV.
(6) Fracture analysis: The fracture surfaces exhibited a quasi-cleavage mechanism, with cleavage facet sizes increasing with distance from the gate. EDS analysis identified calcium oxide inclusions as the primary detrimental non-metallic inclusions, emphasizing the importance of melt cleanliness for minimizing casting defects.
The integrated approach presented in this study provides a robust framework for designing and optimizing ductile iron structural components, effectively balancing lightweight design, manufacturing efficiency, and defect control. The research findings demonstrate the significant potential of ductile iron as an alternative material for commercial vehicle structural components, offering lightweight, cost-effective, and reliable solutions for the automotive industry.