In this work, I studied the design, simulation, optimization, and production validation of ductile iron castings for an integral commercial vehicle crossbeam. My objective was to replace a traditional welded steel crossbeam assembly with a single, high-integrity ductile iron casting. I focused on reducing mass, maintaining structural safety, controlling filling and solidification behavior, minimizing gas entrainment and shrinkage porosity, and verifying the final component through chemical, microstructural, and mechanical characterization. The approach combined finite element static analysis, three-dimensional casting simulation, orthogonal experiment design, analysis of variance, and full-scale foundry trials. The results show that ductile iron castings can provide a technically sound and economically attractive route for lightweight commercial vehicle structural components.
1. Background and Motivation
Commercial vehicle lightweighting is more difficult than passenger car lightweighting because many commercial vehicle components must carry high static and fatigue loads. A crossbeam is a core load-bearing element in the frame system. It transfers vertical, lateral, longitudinal, and torsional loads among the rails. If the crossbeam is made only from lightweight alloys, the strength and stiffness requirements are often difficult to satisfy. If it is made from conventional steel, multiple forming, welding, and assembly steps are required, which increases cost, process variation, and potential weak points. Ductile iron castings offer an alternative because they combine castability, high strength, moderate cost, and good design freedom. In particular, pearlitic ductile iron castings can achieve tensile strengths above 800 MPa while retaining useful elongation.
I selected a pearlitic ductile iron grade for the integral crossbeam because the component must resist bending and torsion while remaining manufacturable by sand casting. The target material was a high-strength ductile iron casting similar to QT800-5 in performance. The integrated design reduced the number of parts from seven to one. This change simplified the manufacturing chain, removed welded joints, reduced mass, and improved geometric continuity. However, the same integration also created large thickness transitions, complex internal cavities, and hot spots. These features made the design of the gating system and feeding system critical. Therefore, I did not rely only on experience. I used numerical simulation to analyze filling, solidification, gas entrainment, and shrinkage defects in the ductile iron castings.
2. Material, Melting, and Characterization Methods
2.1 Raw Materials and Melt Preparation
I prepared the ductile iron melt from high-purity pig iron, carbon steel scrap, recycled foundry returns, and small additions of alloying elements. A rare-earth magnesium ferrosilicon alloy was used as the spheroidizing agent, and a silicon-barium alloy plus ferrosilicon was used for inoculation. The melting operation was performed in a medium-frequency induction furnace. The charge was added in a controlled sequence to avoid oxidation and to stabilize carbon and silicon recovery. After complete melting, I removed slag and allowed the melt to homogenize. Alloying additions such as copper, nickel, and molybdenum were introduced to promote pearlite formation and hardenability in the final ductile iron castings.
The spheroidization treatment was performed by a cored-wire feeding method. This method reduced magnesium loss and improved process consistency. The inoculation treatment was divided into in-ladle inoculation and stream inoculation during pouring. The pouring temperature was kept within a controlled window to balance fluidity and shrinkage. I also ensured that the time between spheroidization and pouring was short enough to avoid degradation of the nodular graphite morphology. These steps are essential for high-quality ductile iron castings because graphite shape controls strength, ductility, and fatigue behavior.
2.2 Characterization Techniques
I analyzed the chemical composition of the ductile iron castings using optical emission spectrometry. I prepared metallographic specimens by grinding, polishing, and etching with nital. I observed graphite morphology and matrix structure using an optical microscope and image analysis software. I evaluated spheroidization grade and graphite size according to standard metallographic rating procedures. I also used scanning electron microscopy and energy-dispersive spectroscopy to examine fracture surfaces and inclusion chemistry. Tensile tests were conducted on specimens taken from different locations of the crossbeam. Hardness was measured using a Vickers hardness tester. These methods allowed me to correlate local solidification conditions with final properties in the ductile iron castings.
3. Integrated Structural Design and Static Analysis
3.1 Original Crossbeam Configuration
The original crossbeam was an assembly of seven steel parts, including L-shaped plates, fixed seats, and a U-shaped member. The total mass was approximately 50.5 kg. The assembly required multiple forming and joining operations. The joined structure had stress concentrations at bolt holes and welded connections. I evaluated the original design under four representative load cases: acceleration, cornering, torsion, and emergency braking. The maximum equivalent stress in the original crossbeam reached about 680 MPa under torsion, which exceeded the allowable range for the selected steel in that local region. This indicated that the original welded design had limited safety margin and that a more integrated solution was needed.
3.2 Design Rules for the Integral Crossbeam
I applied several design rules for the integral ductile iron casting. First, I minimized wall thickness where possible to reduce mass and avoid large hot spots. Second, I used ribs and transition fillets to distribute stress and improve stiffness. Third, I arranged the section geometry to resist torsion. Fourth, I considered sand core removal, mold filling, and feeding. Fifth, I kept machining allowances only where required. These rules are especially important for ductile iron castings because the solidification mode is mushy, and isolated liquid regions can easily form shrinkage defects.
3.3 Stiffness Equivalence and Wall Thickness
To replace steel with ductile iron while maintaining bending stiffness, I used an equivalent stiffness relationship. For a plate-like section, the bending stiffness is proportional to the product of elastic modulus and the cube of thickness. Therefore, I wrote:
$$E_{sl} t_{sl}^{3}=E_{dl} t_{dl}^{3}$$
Here, \(E_{sl}\) and \(E_{dl}\) are the elastic moduli of steel and ductile iron, respectively, and \(t_{sl}\) and \(t_{dl}\) are the corresponding wall thicknesses. Using the material properties of the selected ductile iron and the original steel, I calculated that a minimum wall thickness of about 8.0 mm was needed for the integral ductile iron castings. This value was used as the lower bound for the main load-bearing walls. I then adjusted local thicknesses according to stress distribution and casting process constraints.
3.4 Integrated Geometry
I changed the main load-bearing section from a fabricated I-beam-like assembly to a more continuous rounded axial section. This change improved torsional rigidity and reduced abrupt stiffness changes. I added reinforcement keys at weak regions. I also integrated support platforms and mounting surfaces into the casting. The maximum envelope size of the integrated crossbeam was approximately 839 mm × 403 mm × 221 mm. The mass was reduced from 50.5 kg to 36 kg, a reduction of about 28.7%. This demonstrated that ductile iron castings can achieve meaningful lightweighting through integration and topology improvement, not only through material substitution.
3.5 Load Cases and Safety Factor
I performed static finite element analysis on the integrated crossbeam in the frame environment. I used shell elements for the rails and neighboring plate structures, and tetrahedral elements for the complex ductile iron casting. Bolt and rivet connections were represented by rigid elements. The load cases and maximum equivalent stresses are summarized in Table 1.
| Load Case | Applied Load | Maximum Equivalent Stress in Original Design | Maximum Equivalent Stress in Integrated Ductile Iron Casting |
|---|---|---|---|
| Acceleration | 60,000 N along longitudinal direction | 85.1 MPa | 69.3 MPa |
| Cornering | 80,000 N along lateral direction | 396.1 MPa | 311.1 MPa |
| Torsion | 80,000 N·m equivalent torque | 680.6 MPa | 558.4 MPa |
| Emergency braking | 20,000 N opposing brake force | 95.4 MPa | 95.4 MPa |
The maximum stress in the integrated ductile iron castings occurred under torsional loading at threaded holes. The peak value was 558.4 MPa. For the selected high-strength ductile iron, the allowable strength was taken as 800 MPa. The safety factor was therefore:
$$n=\frac{\sigma_{allow}}{\sigma_{max}}=\frac{800}{558.4}\approx 1.43$$
This safety factor is greater than 1.4, which satisfies the design requirement for the commercial vehicle crossbeam. The integrated ductile iron castings therefore met the structural strength target while reducing mass and part count.
4. Casting Process Design
4.1 Parting Plane and Mold Layout
I selected the parting plane at the maximum projected contour of the crossbeam. This choice simplified mold making, allowed reliable core placement, and reduced the projected area on the parting plane. I used a horizontal sand mold with a self-hardening resin sand core for the internal cavity. The mold box size was 1100 mm × 900 mm × 350 mm. I arranged a two-cavity layout for initial evaluation, but I analyzed a single casting for most simulation cases to reduce computation time. The single-cavity model preserved the thermal and flow conditions relevant to the ductile iron castings.
4.2 Sand Core and Venting
The internal cavity of the crossbeam required a complex sand core. Because the core is surrounded by liquid metal, it must have low gas evolution, sufficient strength, and effective venting. I designed the core with internal vent channels and used resin-bonded sand. The venting path allowed decomposition gases to escape through the core prints and mold vents. If gas cannot escape, it can be trapped in the ductile iron castings and form porosity. Therefore, core venting was treated as a critical process variable.
4.3 Gating System Design
I designed a closed gating system to fill the large-span crossbeam rapidly and smoothly. The cross-sectional area ratio was set as:
$$F_{\text{sprue}}:F_{\text{runner}}:F_{\text{ingate}}=2.50:1.25:1$$
Four ingates were used, each with a trapezoidal cross-section. The total ingate area was calculated from the required volume flow rate:
$$A_{g}=\frac{V}{v_{g}t}$$
Here, \(A_g\) is the total ingate area, \(V\) is the metal volume passing through the ingates, \(v_g\) is the flow velocity, and \(t\) is the filling time. The calculated total ingate area was about 1200 mm². The runner area was about 1500 mm², and the sprue area was about 3000 mm². The pouring time was estimated using a common empirical relationship for cast iron:
$$t=f\left(\sqrt{G}+\sqrt[3]{0.2WG}\right)$$
where \(G\) is the casting mass, \(W\) is the wall thickness, and \(f\) is a material factor. For the integrated crossbeam, the calculated pouring time was about 7.27 s. This corresponded to a nominal pouring speed of about 11 cm/s in the simulation setup. The gating system was intended to provide simultaneous filling through four ingates, reduce turbulent impact, and establish a favorable temperature field for directional solidification.
4.4 Riser Design
Initial simulation showed that isolated liquid regions formed at the two ends of the crossbeam. These regions had thick sections and complex geometry, so they solidified late and could not be fed through the thin central web. I therefore increased the number of risers from four to eight. Four risers on each side were positioned near the thick end regions. The riser size was \(4\times \phi 70\,\text{mm}\times 120\,\text{mm}\) per side in the optimized layout. The purpose of the risers was to supply liquid metal during solidification and to move shrinkage defects away from the load-bearing crossbeam body. This is a standard but critical strategy for sound ductile iron castings.
4.5 Simulation Setup
I imported the three-dimensional models into the casting simulation software. The mesh size was set to 2 mm after a mesh sensitivity check. The total mesh count was approximately \(1.23\times 10^{7}\) elements. The mold material was furan resin sand. The interfacial heat transfer coefficient between the metal and the sand mold was set to 700 W/(m²·K). Gravity was applied in the negative Z direction. The thermophysical properties of the ductile iron were defined as functions of temperature. The liquidus and solidus temperatures were approximately 1457 K and 1418 K, respectively. The latent heat was 210 kJ/kg. These settings allowed me to simulate filling, solidification, gas entrainment, and shrinkage porosity in the ductile iron castings.
| Property | Value |
|---|---|
| Density | 6459 kg/m³ |
| Liquidus temperature | 1457 K |
| Solidus temperature | 1418 K |
| Latent heat | 210 kJ/kg |
| Specific heat capacity | 0.78 kJ/(kg·K) |
5. Simulation Results and Process Optimization
5.1 Initial Filling and Solidification Behavior
In the initial design, the metal entered the mold through the sprue and runner, then divided into four streams through the ingates. The filling sequence was stable at the beginning. The liquid metal filled the bottom of the cavity and then advanced toward the center and ends. The temperature field was relatively uniform except in the thin central web, which cooled faster because of its smaller thermal mass. The filling process completed without obvious misruns. However, the solidification simulation showed that the thick end regions remained liquid for a long time. These regions became isolated from the feeding path. The maximum solidification time at the ends was about 436 s in the initial design. The last liquid regions were detached from the risers, so shrinkage porosity was likely.
The initial predicted shrinkage porosity volume was 751.16 mm³. The defects were concentrated at the two ends of the crossbeam. This result was unacceptable because those regions carry high bending and torsional loads. I therefore modified the riser layout and repeated the simulation. After adding end risers, the maximum solidification time at the ends decreased to about 307.67 s. The isolated liquid regions were eliminated from the load-bearing body. The predicted shrinkage porosity volume decreased to 685.85 mm³, a reduction of 8.69%. More importantly, most remaining porosity was located in the risers and gating system rather than in the crossbeam body.
| Design | Riser Count | Maximum End Solidification Time | Shrinkage Porosity Volume | Defect Location |
|---|---|---|---|---|
| Initial | 4 | 436 s | 751.16 mm³ | Crossbeam ends |
| Optimized | 8 | 307.67 s | 685.85 mm³ | Riser and gating system |
5.2 Effect of Pouring Temperature
I studied three pouring temperatures: 1380 °C, 1400 °C, and 1420 °C. The pouring speed was kept at 11 cm/s, and the mold temperature was 20 °C. The filling behavior was similar at all three temperatures, but local flow details differed. At 1380 °C, the melt was less fluid, and the filling of complex end regions was slower. At 1400 °C, the flow was stable and the cavity filled uniformly. At 1420 °C, the melt filled faster, but local turbulence increased, especially where the ingate streams impinged on the cavity wall. Higher temperature also reduced surface tension and gas solubility, which increased the tendency for gas entrainment.
The gas entrainment volumes were 1700.58 mm³ at 1380 °C, 1682.44 mm³ at 1400 °C, and 1697.82 mm³ at 1420 °C. The minimum occurred at 1400 °C. The shrinkage porosity volumes were 667.92 mm³ at 1380 °C, 685.85 mm³ at 1400 °C, and 702.11 mm³ at 1420 °C. The shrinkage porosity increased with pouring temperature. This is because a higher pouring temperature increases superheat, extends the solidification time, enlarges the mushy zone, and increases volumetric shrinkage. Therefore, lower pouring temperature is beneficial for reducing shrinkage porosity, but too low a temperature can impair filling and increase gas entrainment in ductile iron castings.
| Pouring Temperature | Gas Entrainment Volume | Shrinkage Porosity Volume | Solidification Time |
|---|---|---|---|
| 1380 °C | 1700.58 mm³ | 667.92 mm³ | 307.67 s |
| 1400 °C | 1682.44 mm³ | 685.85 mm³ | 326.96 s |
| 1420 °C | 1697.82 mm³ | 702.11 mm³ | 346.33 s |
5.3 Effect of Pouring Speed
I also studied three pouring speeds: 10 cm/s, 11 cm/s, and 12 cm/s. The pouring temperature was 1400 °C, and the mold temperature was 20 °C. At 10 cm/s, the filling front was smooth and stable, but the total filling time was longer, so the thin sections cooled more before the cavity was completely filled. At 11 cm/s, the filling process was balanced. At 12 cm/s, the filling time decreased, but local splashing and vortex formation became more noticeable. The faster flow increased the kinetic energy of the melt and promoted gas entrainment.
The gas entrainment volumes were 1655.77 mm³ at 10 cm/s, 1682.44 mm³ at 11 cm/s, and 1700.39 mm³ at 12 cm/s. The gas entrainment increased monotonically with pouring speed. The shrinkage porosity volumes were 696.92 mm³ at 10 cm/s, 685.85 mm³ at 11 cm/s, and 680.85 mm³ at 12 cm/s. The shrinkage porosity decreased as pouring speed increased because the mold filled more quickly and the temperature field remained more uniform. However, faster filling also increased the risk of gas entrainment. Thus, pouring speed must be optimized together with pouring temperature in ductile iron castings.
| Pouring Speed | Gas Entrainment Volume | Shrinkage Porosity Volume |
|---|---|---|
| 10 cm/s | 1655.77 mm³ | 696.92 mm³ |
| 11 cm/s | 1682.44 mm³ | 685.85 mm³ |
| 12 cm/s | 1700.39 mm³ | 680.85 mm³ |
5.4 Orthogonal Experiment and Analysis of Variance
To identify the best combination of pouring speed and pouring temperature, I designed a two-factor, three-level orthogonal experiment. The factors were pouring speed \(A\) and pouring temperature \(B\). The levels were 10, 11, and 12 cm/s for \(A\), and 1380, 1400, and 1420 °C for \(B\). The response variables were gas entrainment volume \(Y_g\) and shrinkage porosity volume \(Y_s\). I used analysis of variance to determine which factors had statistically significant effects. The \(F\) value for each factor was calculated as:
$$F=\frac{MS_{\text{factor}}}{MS_{\text{error}}}$$
where \(MS_{\text{factor}}\) is the mean square of the factor and \(MS_{\text{error}}\) is the mean square of the error. A small \(p\)-value indicates a statistically significant effect. The experimental layout and results are shown in Table 4.
| Run | Pouring Speed \(A\) | Pouring Temperature \(B\) | Gas Entrainment Volume | Shrinkage Porosity Volume |
|---|---|---|---|---|
| 1 | 10 cm/s | 1380 °C | 1677.19 mm³ | 679.18 mm³ |
| 2 | 10 cm/s | 1400 °C | 1655.77 mm³ | 696.92 mm³ |
| 3 | 10 cm/s | 1420 °C | 1668.84 mm³ | 713.20 mm³ |
| 4 | 11 cm/s | 1400 °C | 1682.44 mm³ | 685.85 mm³ |
| 5 | 11 cm/s | 1420 °C | 1697.82 mm³ | 702.11 mm³ |
| 6 | 11 cm/s | 1380 °C | 1700.58 mm³ | 667.92 mm³ |
| 7 | 12 cm/s | 1420 °C | 1704.32 mm³ | 697.20 mm³ |
| 8 | 12 cm/s | 1400 °C | 1700.39 mm³ | 680.85 mm³ |
| 9 | 12 cm/s | 1380 °C | 1721.51 mm³ | 663.00 mm³ |
For gas entrainment, the range of pouring speed was 124.42, while the range of pouring temperature was 60.68. This indicates that pouring speed had a stronger influence on gas entrainment than pouring temperature. The analysis of variance for gas entrainment is shown in Table 5.
| Source | Sum of Squares | Degrees of Freedom | Mean Square | \(F\) | \(p\) |
|---|---|---|---|---|---|
| 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 | — | — |
For shrinkage porosity, the range of pouring speed was 48.29, while the range of pouring temperature was 102.45. This indicates that pouring temperature had a stronger influence on shrinkage porosity than pouring speed. The analysis of variance for shrinkage porosity is shown in Table 6.
| Source | Sum of Squares | Degrees of Freedom | Mean Square | \(F\) | \(p\) |
|---|---|---|---|---|---|
| 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 | — | — |
The orthogonal analysis showed that pouring speed significantly affected gas entrainment, while pouring temperature significantly affected shrinkage porosity. Based on the combined responses, I selected the best compromise as a pouring speed of 11 cm/s and a pouring temperature of 1380 °C, with a mold temperature of 20 °C. This combination reduced shrinkage porosity while keeping gas entrainment at an acceptable level. In the optimized simulation, the shrinkage porosity volume was 667.92 mm³, which was 11.74% lower than the initial design. The filling front remained stable, and the final liquid regions were confined to the risers and gating system. The optimized process therefore provided a reliable basis for producing sound ductile iron castings.
6. Production Trial and Experimental Validation
6.1 Melting and Pouring Trial
I conducted a full-scale production trial using the optimized gating system and process parameters. The pouring temperature was 1380 °C, the pouring speed was approximately 10–11 cm/s, and the mold temperature was about 20 °C. The spheroidization and inoculation treatments were performed in the same manner as the laboratory trials. After pouring, the castings were allowed to cool in the mold, then shaken out, cleaned, and inspected. The risers and gating system were removed by cutting and grinding. The casting surfaces were inspected for visible defects such as sand burn-on, cold shuts, and shrinkage depressions. No obvious surface defects were found in the load-bearing regions.
6.2 Chemical Composition
I measured the chemical composition of the final ductile iron castings at three locations and averaged the results. The composition is shown in Table 7. The carbon equivalent was calculated as:
$$CE=C+\frac{Si}{3}+\frac{P}{2}$$
The carbon equivalent was close to the eutectic range, which is favorable for castability and feeding. The residual magnesium content was sufficient for nodular graphite formation. The phosphorus and sulfur contents were low, which reduced the risk of phosphide eutectic and sulfide inclusions. Alloying elements such as copper, nickel, and molybdenum were controlled to promote a pearlitic matrix. These results confirm that the melt chemistry was suitable for high-strength ductile iron castings.
| 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 |
6.3 Microstructure
I sampled three regions of the crossbeam: near the ingate, at the thin central web, and at the far end. The graphite morphology and matrix structure were evaluated. The spheroidization grade was grade 2 in all regions. The graphite size was grade 6–7. The nodule counts were approximately 180 nodules/mm² near the ingate, 252 nodules/mm² at the thin central web, and 212 nodules/mm² at the far end. The matrix was predominantly pearlitic, with pearlite content above 90%. Carbide content was about 1% or less, and no phosphide eutectic was observed. These microstructural features are typical of high-quality ductile iron castings with good strength and moderate ductility.
| Sampling Location | Spheroidization Grade | Graphite Size Grade | Nodule Count | Pearlite Content | Carbide Content | Phosphide Eutectic |
|---|---|---|---|---|---|---|
| Near ingate | 2 | 6 | 180 /mm² | ≥90% | ≤1% | None |
| Thin central web | 2 | 7 | 252 /mm² | ≥95% | ≤1% | None |
| Far end | 2 | 6 | 212 /mm² | ≥90% | ≤1% | None |
6.4 Mechanical Properties
I performed tensile tests on specimens taken from the three sampling regions. The results are shown in Table 9. The near-ingate region had the highest elongation, 7.6%, with a yield strength of 546 MPa and a tensile strength of 875 MPa. The thin central web had the highest tensile strength, 891 MPa, with a yield strength of 562 MPa and an elongation of 6.0%. The far end had a tensile strength of 854 MPa, a yield strength of 552 MPa, and an elongation of 6.5%. All regions met the target requirements for the high-strength ductile iron castings. The hardness values were also similar among the three regions, all above 270 HV. These results indicate that the optimized process produced uniform and reliable mechanical properties.
| Sampling Location | Yield Strength \(R_{p0.2}\) | Tensile Strength \(R_m\) | Elongation \(A\) | Hardness |
|---|---|---|---|---|
| Near ingate | 546 MPa | 875 MPa | 7.6% | 284.1 HV |
| Thin central web | 562 MPa | 891 MPa | 6.0% | 281.9 HV |
| Far end | 552 MPa | 854 MPa | 6.5% | 285.9 HV |
6.5 Fracture and Inclusion Analysis
I examined the tensile fracture surfaces using scanning electron microscopy. The fracture surfaces contained cleavage steps and river patterns around the graphite nodules. These features indicate a mixed fracture mode with limited ductility, which is consistent with a high-strength pearlitic ductile iron. The near-ingate specimen showed finer cleavage platforms and fewer river patterns, which correlated with its higher elongation. The far-end specimen showed larger cleavage platforms and more river patterns, indicating lower local ductility. This variation is related to the local cooling rate and the distance from the ingate.
Energy-dispersive spectroscopy revealed that some inclusions were rich in calcium and oxygen. These inclusions were identified as calcium oxide. The sharp morphology of these oxide inclusions can cause stress concentration and act as crack initiation sites. Compared with graphite nodules, which are relatively rounded, sharp oxide inclusions are more detrimental to the mechanical properties of ductile iron castings. Therefore, melt cleanliness and slag removal are essential to ensure high performance. The optimized process reduced the inclusion content, but further improvements in filtration and slag control could be beneficial.
7. Discussion
The results show that the integral crossbeam design and the optimized casting process are mutually dependent. The structural integration reduced mass and part count, but it also created thick end regions that required additional feeding. Without riser optimization, the ductile iron castings would have contained shrinkage porosity in load-bearing areas. By using numerical simulation, I identified the last solidifying regions and repositioned the risers. This approach reduced the shrinkage porosity volume and moved the remaining defects into the risers and gating system. The optimized process also controlled gas entrainment by balancing pouring speed and pouring temperature.
The orthogonal experiment showed that pouring speed and pouring temperature have different effects on different defects. Pouring speed mainly controls gas entrainment because it governs turbulence, surface disturbance, and bubble entrapment. Pouring temperature mainly controls shrinkage porosity because it governs superheat, thermal gradient, and solidification time. Therefore, a single-parameter optimization is not sufficient for complex ductile iron castings. A multi-objective approach is necessary. In my work, the best compromise was 11 cm/s pouring speed and 1380 °C pouring temperature. This combination kept gas entrainment low enough while minimizing shrinkage porosity.
The production trial confirmed the simulation results. The chemical composition was stable, the graphite was well nodularized, and the matrix was predominantly pearlitic. The tensile properties exceeded the target values in all sampled regions. The hardness was uniform. The fracture analysis showed that the main residual risk is oxide inclusion content. This is a common challenge in ductile iron castings because magnesium treatment can produce reaction products that must be removed before pouring. Good slag removal, ceramic filters, and controlled pouring practices can further improve cleanliness.
8. Practical Implications
From an engineering standpoint, my study provides a practical route for commercial vehicle lightweighting using ductile iron castings. The integrated design reduced the mass of the crossbeam from 50.5 kg to 36 kg, a 28.7% reduction. The static safety factor was about 1.43, which meets the design requirement. The optimized gating and riser system reduced the predicted shrinkage porosity volume from 751.16 mm³ to 685.85 mm³, and the final process reduced it further to 667.92 mm³. The mechanical properties reached a tensile strength above 850 MPa, a yield strength above 540 MPa, and an elongation above 6.0%. These values demonstrate that ductile iron castings can satisfy the load-bearing requirements of commercial vehicle crossbeams.
The process also offers manufacturing advantages. A single casting replaces seven assembled parts. This reduces welding, fixturing, and inspection operations. It also improves dimensional consistency and removes weld-related fatigue risks. However, the foundry must control melting, spheroidization, inoculation, pouring temperature, pouring speed, mold filling, and feeding. Simulation is a powerful tool for this purpose. It allows the engineer to visualize filling and solidification, predict shrinkage porosity and gas entrainment, and compare process alternatives before production. In my work, simulation reduced trial-and-error and provided a reliable basis for the final process.
9. Conclusions
I designed and optimized an integral crossbeam made from high-strength ductile iron castings. The main conclusions are as follows.
1. The integrated crossbeam reduced the number of parts from seven to one and reduced mass from 50.5 kg to 36 kg, a reduction of 28.7%. The maximum equivalent stress under all load cases was 558.4 MPa, and the safety factor was about 1.43. The design satisfies the strength requirement for the commercial vehicle crossbeam.
2. The initial casting design produced isolated liquid regions at the crossbeam ends, leading to a shrinkage porosity volume of 751.16 mm³. By increasing the riser count from four to eight and placing the risers at the ends, the shrinkage porosity volume decreased to 685.85 mm³. The remaining defects were shifted to the risers and gating system.
3. Pouring temperature and pouring speed had different effects on defect formation. The minimum gas entrainment occurred at 1400 °C, while the minimum shrinkage porosity occurred at 1380 °C. The minimum gas entrainment occurred at 10 cm/s, while the minimum shrinkage porosity occurred at 12 cm/s. The orthogonal experiment and analysis of variance showed that pouring speed significantly affected gas entrainment, while pouring temperature significantly affected shrinkage porosity.
4. The optimized process parameters were a pouring speed of 11 cm/s, a pouring temperature of 1380 °C, and a mold temperature of 20 °C. The final shrinkage porosity volume was 667.92 mm³, which was 11.74% lower than the initial design. The filling process was stable, and the final solidification regions were confined to the risers and gating system.
5. The production trial produced sound ductile iron castings. The chemical composition was stable, with low phosphorus and sulfur and sufficient residual magnesium. The graphite spheroidization grade was grade 2, and the graphite size was grade 6–7. The matrix was predominantly pearlitic, with pearlite content above 90%. The tensile strength was above 854 MPa, the yield strength was above 546 MPa, and the elongation was above 6.0%. Hardness was above 270 HV in all sampled regions. These properties meet the target requirements.
6. Fracture analysis showed cleavage steps and river patterns, indicating a high-strength pearlitic ductile iron with limited ductility. Calcium oxide inclusions were observed and can act as crack initiation sites. Therefore, melt cleanliness and slag removal remain important for further improvement. Overall, the optimized process is reliable and suitable for producing high-performance ductile iron castings for commercial vehicle crossbeams.
10. Future Outlook
Further work can extend this study in several directions. First, fatigue testing and full vehicle validation should be performed to evaluate long-term service performance. Second, heat treatment such as austempering could be investigated to improve the strength-toughness balance of ductile iron castings. Third, advanced filtration and melt treatment could further reduce oxide inclusions and improve fracture resistance. Fourth, the simulation model could be coupled with microstructure prediction to optimize nodule count and pearlite content locally. Fifth, the integrated crossbeam concept could be extended to other commercial vehicle frame components. With continued development, ductile iron castings will remain a competitive solution for high-load, safety-critical, and cost-sensitive structural components.