The tractor saddle is a critical connecting component between the tractor and the semi-trailer. During operation, it experiences complex impact loads from multiple directions, requiring excellent mechanical properties and high fatigue resistance. Traditionally, such saddles are manufactured from forged steel, which offers good impact toughness but also incurs high material and processing costs. To reduce production cost while maintaining the required service performance, the foundry industry has attempted to replace forged steel saddles with cast iron saddles, specifically ductile iron castings. However, the casting process for such saddle components is prone to casting defect formation, including choke burn, cold shut, shrinkage porosity, shrinkage cavity, sand inclusion and slag inclusion. These casting defects cause a large number of rejected products and significant economic losses. In this context, the present study focuses on the optimization of the sand casting process for a tractor saddle by means of systematic design, numerical simulation, orthogonal experimental optimization and experimental verification. The objective is to minimize casting defect probability and to achieve a reliable “iron instead of steel” solution for the tractor saddle.
In this study, I adopt a comprehensive approach combining structural analysis, gating system design, computational simulation and experimental validation. The objective is to reduce casting defects, especially shrinkage porosity and shrinkage cavity, in a ductile iron tractor saddle. The research starts with the analysis of the structural processability of the saddle. The saddle is a large thin-wall casting with an envelope dimension of 755 mm × 580 mm × 322 mm and a net weight of 63 kg. The average wall thickness is 10 mm, with a maximum thickness of 30 mm. Because the saddle is a typical large thin-wall casting, the filling and solidification processes are challenging. The proper selection of gating system, riser system, filter system and venting system is crucial to avoid casting defects.
Structural Analysis and Process Design
The tractor saddle is designed as a ductile iron casting with the material grade QT700-6. This material belongs to the pearlitic ductile iron family, offering high strength, good wear resistance and moderate ductility. The chemical composition of QT700-6 is carefully controlled to balance strength and castability. Table 1 shows the main chemical composition limits used for the saddle material.
| C | Si | Mn | S | P | Mg | Ce | Cr | Cu | Mo |
|---|---|---|---|---|---|---|---|---|---|
| 3.5–3.9 | 1.8–2.1 | 0.35–0.5 | 0.006–0.02 | ≤0.05 | 0.035–0.055 | 0.006–0.03 | ≤0.1 | 0.4–0.6 | ≤0.15 |
The mechanical property requirements for the cast saddle are: tensile strength ≥ 700 MPa, elongation ≥ 6%, and hardness between 240 and 290 HBW. According to the DIN EN ISO 945 standard, the graphite in the casting should be type VI (spheroidal graphite), with a graphite size of 5–8 grade. No flake graphite is allowed. The matrix should be predominantly pearlitic with a pearlite content of at least 80% and ferrite content not exceeding 20%. These requirements ensure that the cast iron saddle can replace the forged steel saddle without compromising the structural integrity under service loads.
The molding process was determined based on the geometry and production conditions. A two-box molding method was selected, with a horizontal parting line. Compared with vertical gating, horizontal casting minimizes the height difference and simplifies the core structure. The casting process parameters include a shrinkage allowance of 1.1%, a draft angle of 3°, and a dimensional tolerance grade of CT 12 according to GB/T 6414-2017. The machining allowance grade is selected as RMAG F–H. The weight tolerance grade is MT 11–14. These parameters are essential for producing a sound casting with dimensional accuracy and minimal casting defects.
The sand core design is also critical for the saddle casting. The saddle cavity is completely through, so a single cold-box core is used. The core is fixed with vertical and horizontal core prints. The vertical core prints are designed with dimensions of 108 mm × 75 mm × 42 mm for the top and 108 mm × 98 mm × 42 mm for the bottom. The horizontal core prints are symmetrically arranged with a left rear large core print of 64 mm and a left front small core print of 35 mm. The draft angle for the core prints is 3°. This core design ensures accurate positioning, adequate support and good venting during casting, which helps reduce casting defects such as gas porosity and sand inclusion.
Theoretical Analysis of Filling and Solidification
The numerical simulation of the casting process is based on the fundamental conservation laws of mass, momentum and energy. In the filling stage, the molten metal is treated as an incompressible fluid. The governing equations are expressed as follows.
The continuity (mass conservation) equation is:
$$
\frac{\partial \rho}{\partial t} + \frac{\partial (\rho u)}{\partial x} + \frac{\partial (\rho v)}{\partial y} + \frac{\partial (\rho w)}{\partial z} = 0
$$
For steady-state incompressible flow, it reduces to:
$$
\frac{\partial u}{\partial x} + \frac{\partial v}{\partial y} + \frac{\partial w}{\partial z} = 0
$$
The momentum conservation equations (Navier-Stokes equations) in the x, y, and z directions are:
$$
\rho \left( \frac{\partial u}{\partial t} + u \frac{\partial u}{\partial x} + v \frac{\partial u}{\partial y} + w \frac{\partial u}{\partial z} \right) = -\frac{\partial p}{\partial x} + \mu \nabla^2 u + \rho g_x
$$
$$
\rho \left( \frac{\partial v}{\partial t} + u \frac{\partial v}{\partial x} + v \frac{\partial v}{\partial y} + w \frac{\partial v}{\partial z} \right) = -\frac{\partial p}{\partial y} + \mu \nabla^2 v + \rho g_y
$$
$$
\rho \left( \frac{\partial w}{\partial t} + u \frac{\partial w}{\partial x} + v \frac{\partial w}{\partial y} + w \frac{\partial w}{\partial z} \right) = -\frac{\partial p}{\partial z} + \mu \nabla^2 w + \rho g_z
$$
The energy conservation equation for the filling and solidification process is:
$$
\rho C_p \left( \frac{\partial T}{\partial t} + u \frac{\partial T}{\partial x} + v \frac{\partial T}{\partial y} + w \frac{\partial T}{\partial z} \right) = \frac{\partial}{\partial x} \left( k \frac{\partial T}{\partial x} \right) + \frac{\partial}{\partial y} \left( k \frac{\partial T}{\partial y} \right) + \frac{\partial}{\partial z} \left( k \frac{\partial T}{\partial z} \right) + Q
$$
In the above equations, $\rho$ is the density, $u$, $v$, $w$ are the velocity components, $p$ is the pressure, $\mu$ is the dynamic viscosity, $C_p$ is the specific heat capacity, $T$ is the temperature, $k$ is the thermal conductivity, and $Q$ is the latent heat source term. During solidification, the latent heat is released and can be expressed as:
$$
Q = \rho L \frac{\partial f_s}{\partial t}
$$
where $L$ is the latent heat of fusion and $f_s$ is the solid fraction. The relationship between temperature and solid fraction can be approximated by a linear or quadratic function. For the ductile iron saddle, the solidification process is strongly affected by the graphite expansion, which influences the shrinkage porosity formation. Therefore, accurate prediction of casting defects such as shrinkage porosity and shrinkage cavity requires a coupled simulation of flow, temperature and solidification.
Design of the Gating System
Based on the structural characteristics and production requirements, a closed gating system was adopted for the tractor saddle. The closed gating system has a ratio of cross-sectional areas of the inner runner (∑Ainner), cross runner (∑Across) and sprue (∑Asprue) equal to 1:1.65:1.9. This ratio helps to keep the molten metal flow stable and to trap slag and impurities, thereby reducing casting defects such as slag inclusion and sand inclusion. The gating system is primarily bottom-fed, combined with a stratified injection mode. This design ensures smooth filling, reduces turbulence and minimizes gas entrapment.
The filling time is an important parameter. For an iron casting weighing less than 450 kg, the filling time is calculated by the empirical formula:
$$
t = S \sqrt{G}
$$
where $G$ is the total weight of the casting including the gating system, taken as 157 kg, and $S$ is a coefficient depending on the wall thickness. For a wall thickness of 8–15 mm, $S = 1.85$. Thus, the filling time is:
$$
t = 1.85 \times \sqrt{157} \approx 23.2 \ \text{s}
$$
However, considering the actual mold condition and the need for stable filling, a filling time of 14 s was selected in the simulation and production. This value was obtained by adjusting the coefficient to 1.85 with a reduced weight because the actual flow condition is modified by the gating system. In the simulation, the filling time was set to 14 s.
The average pressure head $H_p$ is calculated from the geometry of the gating system. For a bottom-gated system, the formula is:
$$
H_p = H_0 – \frac{h_0^2}{2h}
$$
where $H_0$ is the height from the sprue top to the inner runner (300 mm), $h_0$ is the height of the casting above the inner runner (320 mm), and $h$ is the total casting height (320 mm). Substituting the values gives:
$$
H_p = 300 – \frac{320^2}{2 \times 320} = 300 – 160 = 140 \ \text{mm}
$$
The minimum residual pressure head $H_M$ is required to ensure complete filling of the highest and farthest point of the casting. It is calculated by:
$$
H_M \ge L \tan \alpha
$$
where $L$ is the flow length from the sprue center to the farthest point, taken as 195 mm, and $\alpha$ is the pressure angle, taken as 8°. Thus:
$$
H_M \ge 195 \times \tan 8^\circ \approx 27 \ \text{mm}
$$
The total cross-sectional area of the inner runners is determined using the hydraulic formula:
$$
\sum A_{\text{inner}} = \frac{G}{\mu t \sqrt{H_p}}
$$
with $G = 157$ kg, $t = 14$ s, $\mu = 0.64$, and $H_p = 14$ cm. The resulting total area is:
$$
\sum A_{\text{inner}} = \frac{157}{0.64 \times 14 \times \sqrt{14}} \approx 15.6 \ \text{cm}^2
$$
Based on this total area, three different gating systems with 4, 6, and 8 inner runners were designed. Table 2 summarizes the dimensions of the three systems.
| Number of inner runners | Single inner runner area (cm²) | Side length of inner runner (cm) | Cross runner area (cm²) | Sprue area (cm²) | Sprue diameter (cm) |
|---|---|---|---|---|---|
| 4 | 3.9 | 2.0 | 25.74 | 29.64 | 6.0 |
| 6 | 2.6 | 1.6 | 25.74 | 29.64 | 6.0 |
| 8 | 1.95 | 1.4 | 25.74 | 29.64 | 6.0 |
Design of Filtering and Exhaust Systems
To further reduce casting defects, a multi-stage filtering system was designed. The filtering system consists of a choke plate (flow restrictor) at the exit of the sprue, a ceramic foam filter (90 × 90 × 22 mm, 10 ppi) in the front part of the cross runner, and a slag pocket at the end of the cross runner. This combined filtering approach effectively removes slag, dross and other impurities from the molten metal, thus minimizing slag inclusion and sand inclusion defects.
The exhaust system design is essential to allow air and gas to escape from the mold cavity and sand cores. For a large thin-wall casting, the total exhaust area should be 1.5–2.5 times the choke area of the gating system. The venting holes are placed at the highest point of the casting and at the end of the runner system. According to the calculation using the formula:
$$
S = (1.5 \sim 4) \times \frac{G}{t \rho \mu h_p}
$$
with $G = 157$ kg, $\rho = 6.984$ g/cm³, $h_p = 140$ mm, $t = 14$ s and $\mu = 0.35$, the required total exhaust area is approximately 11 cm². The final exhaust design includes one large vent at the runner end with an area of 8 cm² and two small vents at the highest point with an area of 1.5 cm² each. The vents are arranged along the parting line to ensure smooth gas release. For the sand core, venting is provided through the core prints, and the core is made from a low gas evolution, high permeability cold-box sand to prevent gas-related casting defects.
Numerical Simulation of Different Gating Systems
The AnyCasting simulation software was used to model the filling, solidification and defect formation processes of the saddle casting. The initial conditions were set as follows: pouring temperature of 1395 °C, filling time of 14 s, gravity along the +x direction, and natural cooling in the sand mold. The thermophysical properties of FCD 700 (QT700-6) were defined as temperature-dependent functions. The meshing of the casting, cores and gating system was performed with different mesh densities. The core was meshed with a cell size of 2.2 mm³, the runner system with 4.5 mm³, and the casting cavity with 9.1 mm³.
The filling behavior of the molten metal was simulated for the three gating systems. The velocity field during filling indicates the stability of the flow. For the 4-runner system, the molten metal entered the cavity with significant velocity fluctuations. At a fill fraction of 35%, the maximum velocity reached 157 mm/s while the minimum was only 0.699 mm/s, indicating strong turbulence and possible air entrapment. This turbulence can lead to gas porosity and oxide film entrapment, which are typical casting defects. At 55% fill, the maximum velocity increased to 179 mm/s, further aggravating the flow instability. Therefore, the 4-runner design was considered unsuitable for producing a sound saddle casting.
For the 6-runner system, the velocity field was much more uniform. During the entire filling process, the molten metal flowed smoothly without drastic velocity differences. The maximum velocity remained below 120 mm/s and the filling was stable. The cavity was completely filled without obvious turbulence. This smooth flow minimizes the risk of gas entrapment and erosion of the sand mold, thus reducing casting defects.
For the 8-runner system, the flow was stable in the early and middle stages, but at the later stage (about 95% fill) the velocity increased slightly. Nevertheless, the overall flow behavior was acceptable. However, the solidification analysis revealed some concerns, as discussed next.
The solidification sequence and temperature field were analyzed for each design. The formation of isolated liquid regions during solidification is a direct indicator of potential shrinkage porosity and shrinkage cavity. For the 4-runner system, at a solidification fraction of 66.2%, several isolated liquid regions appeared, but they were relatively small and dispersed. At 85.9% solidification, the isolated regions remained but were not very large. For the 6-runner system, the solidification proceeded more uniformly. At 91.9% solidification, only a few small isolated liquid regions remained. This indicates a low risk of shrinkage-related casting defects. For the 8-runner system, elongated isolated liquid regions appeared near the cavity edges at 66.5% solidification. At 85.4%, multiple large and dispersed isolated liquid regions remained, which can cause internal cavities and lead to shrinkage porosity and shrinkage cavity defects. Therefore, the 8-runner system was considered less favorable.
The defect probability parameters were also evaluated. The software uses a temperature-gradient-based criterion combined with local solidification time and feeding resistance. Figure 4 in the original thesis showed the defect probability plots for the 4-, 6- and 8-runner systems. The 4-runner system displayed line-like defects on the upper surface, numerous defect spots in the main cavity, and line-like defects at the bottom. The 6-runner system showed only minor line-like defects near the locking hole at the bottom of the main cavity, with no obvious defects elsewhere. The 8-runner system exhibited red line-like defects on the upper surface, dense defect regions in the main cavity, and some line-like defects at the bottom. Based on these simulations, the 6-runner gating system was selected as the optimal design because it produced the fewest casting defects and the most stable filling and solidification behavior.
Orthogonal Experimental Optimization
To further reduce the casting defect probability, an orthogonal experiment was designed based on the 6-runner gating system. Four factors were selected as variables: pouring temperature (A), filling pressure (B), pouring velocity (C), and molten metal weight (D). Each factor had three levels, as shown in Table 3. The orthogonal array L9(3⁴) was used, resulting in nine simulation trials. The evaluation indices were the total pore volume (cm³) and the solidification time (h). A smaller pore volume indicates fewer shrinkage porosity and shrinkage cavity defects. A shorter solidification time indicates higher productivity.
| Level | Pouring temperature (°C) | Filling pressure (Pa) | Pouring velocity (m/s) | Molten metal weight (kg) |
|---|---|---|---|---|
| 1 | 1395 | 1.0 | 0.34 | 157 |
| 2 | 1385 | 0.9 | 0.30 | 155 |
| 3 | 1405 | 1.1 | 0.40 | 159 |
The L9 orthogonal test scheme and the corresponding simulation results are listed in Table 4.
| Trial | A (°C) | B (Pa) | C (m/s) | D (kg) | Pore volume (cm³) | Solidification time (h) |
|---|---|---|---|---|---|---|
| 1 | 1395 | 1.0 | 0.34 | 157 | 0.986 | 0.5575 |
| 2 | 1395 | 0.9 | 0.30 | 155 | 0.654 | 0.5581 |
| 3 | 1395 | 1.1 | 0.40 | 159 | 0.473 | 0.5542 |
| 4 | 1385 | 1.0 | 0.30 | 159 | 0.813 | 0.5458 |
| 5 | 1385 | 0.9 | 0.40 | 157 | 0.634 | 0.5411 |
| 6 | 1385 | 1.1 | 0.34 | 155 | 0.746 | 0.5417 |
| 7 | 1405 | 1.0 | 0.40 | 155 | 0.682 | 0.5708 |
| 8 | 1405 | 0.9 | 0.34 | 159 | 0.914 | 0.5728 |
| 9 | 1405 | 1.1 | 0.30 | 157 | 0.753 | 0.5739 |
The range analysis (extreme difference analysis) was performed for both indices. The mean values of each level, denoted as $K_1$, $K_2$, $K_3$, and the range $R$ for each factor are summarized in Table 5.
| Index | A | B | C | D | |
|---|---|---|---|---|---|
| Pore volume | K₁ | 0.704 | 0.827 | 0.882 | 0.791 |
| K₂ | 0.731 | 0.734 | 0.740 | 0.694 | |
| K₃ | 0.783 | 0.657 | 0.596 | 0.733 | |
| R | 0.079 | 0.170 | 0.286 | 0.097 | |
| Optimal level | A1 | B3 | C3 | D2 | |
| Solidification time | K₁ | 0.5566 | 0.5580 | 0.5573 | 0.5575 |
| K₂ | 0.5429 | 0.5573 | 0.5593 | 0.5569 | |
| K₃ | 0.5725 | 0.5566 | 0.5554 | 0.5576 | |
| R | 0.0296 | 0.0014 | 0.0039 | 0.0007 | |
| Optimal level | A2 | B3 | C3 | D2 |
From Table 5, when the pore volume is the sole index, the order of factors affecting the result is C > B > D > A, i.e., pouring velocity has the strongest influence on shrinkage porosity and shrinkage cavity. The optimal combination is A1B3C3D2: pouring temperature 1395 °C, filling pressure 1.1 Pa, pouring velocity 0.40 m/s, and molten metal weight 155 kg. When the solidification time is the sole index, the order is A > C > B > D, and the optimal combination is A2B3C3D2: pouring temperature 1385 °C, filling pressure 1.1 Pa, pouring velocity 0.40 m/s, and molten metal weight 155 kg. The two single-index optima differ in pouring temperature. To resolve this conflict, a matrix analysis method was employed.
The matrix analysis assigns weights to each factor and level to yield a global optimal combination. Using the methods described in Section 5.2.2, the weighted relative importance values for each factor level were calculated. The maximum value for each factor corresponds to the best level. The results are summarized in Table 6.
| A1 | A2 | A3 | B1 | B2 | B3 | C1 | C2 | C3 | D1 | D2 | D3 | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Weight | 0.1605 | 0.1632 | 0.1545 | 0.0463 | 0.0513 | 0.0566 | 0.0799 | 0.0916 | 0.1095 | 0.0271 | 0.0305 | 0.0290 |
The maximum weight values are A2, B3, C3 and D2. Therefore, the globally optimal combination is A2B3C3D2, i.e., pouring temperature 1385 °C, filling pressure 1.1 Pa, pouring velocity 0.40 m/s, and molten metal weight 155 kg.
Verification of the Optimal Scheme
To verify the optimized process parameters, a numerical simulation was conducted with the optimal combination. The velocity field showed stable filling with no obvious turbulence. The solidification sequence followed the desired directional pattern, and the isolated liquid regions were small and eliminated gradually. The defect probability simulation indicated that the pore volume was reduced to 0.401 cm³, significantly lower than the values in the orthogonal experiments. The line-like defect previously observed near the locking hole disappeared. The optimized simulation confirmed that the process effectively reduces shrinkage porosity and shrinkage cavity defects, as well as other casting defects such as gas porosity and sand inclusion.
The optimized process was then applied to actual production of the tractor saddle castings. The castings were inspected using ultrasonic testing, and no unacceptable defects were found. Two batches totaling 14 samples were taken for metallographic examination and mechanical testing. The samples were etched with nitric acid alcohol to reveal the microstructure. The results are summarized in Table 7.
| Sample No. | Elongation A (%) | Yield strength (MPa) | Tensile strength (MPa) | Hardness (HBW) | Graphite size (grade) | Pearlite content (%) | Spheroidization grade |
|---|---|---|---|---|---|---|---|
| 01K | 10.0 | 424 | 739 | 245 | 6 | 80 | VI |
| 02K | 10.3 | 412 | 725 | 247 | 6 | 81 | VI |
| 03K | 10.7 | 426 | 747 | 248 | 6 | 80 | VI |
| 04K | 10.0 | 411 | 749 | 240 | 6 | 75 | VI |
| 05K | 10.2 | 436 | 728 | 246 | 6 | 80 | VI |
| 06K | 10.1 | 432 | 735 | 249 | 6 | 81 | VI |
| 07K | 10.3 | 416 | 736 | 247 | 6 | 81 | VI |
| 08K | 10.6 | 414 | 745 | 245 | 6 | 80 | VI |
| 09K | 10.5 | 433 | 742 | 247 | 6 | 82 | VI |
| 10K | 10.2 | 421 | 728 | 249 | 6 | 82 | VI |
| 11K | 10.0 | 424 | 726 | 246 | 6 | 81 | VI |
| 12K | 10.1 | 410 | 731 | 248 | 6 | 80 | VI |
| 13K | 10.1 | 433 | 744 | 246 | 6 | 83 | VI |
| 14K | 10.5 | 424 | 742 | 247 | 6 | 82 | VI |
All tested specimens met the specified requirements: elongation ≥ 6%, tensile strength ≥ 700 MPa, hardness in the range of 240–290 HBW, graphite size grade 6, pearlite content ≥ 80% (except sample 04K which was slightly below), and spheroidization grade VI. The production yield of the tractor saddle increased from about 60% in the original process to above 95% after the optimization. This demonstrates that the optimized gating system and process parameters significantly reduce casting defects and improve the overall quality of the saddle casting.
Conclusion
In this study, a systematic approach was applied to optimize the sand casting process of a ductile iron tractor saddle. The following conclusions can be drawn:
- The tractor saddle was successfully redesigned as a QT700-6 ductile iron casting with an envelope size of 755 mm × 580 mm × 322 mm. The two-box horizontal molding method and a single cold-box core were adopted to simplify the casting process and to reduce casting defects.
- A closed gating system with the area ratio ∑Ainner:∑Across:∑Asprue = 1:1.65:1.9 was designed. Three variants with 4, 6, and 8 inner runners were evaluated using AnyCasting numerical simulation. The 6-runner gating system produced the most stable filling, a favorable solidification sequence, and the lowest defect probability. The final gating system had a single sprue with a diameter of 6 cm, one cross runner with a cross-sectional area of 25.74 cm², and six square inner runners each with a side length of 1.6 cm.
- The filtering system combined a flow restrictor, a ceramic foam filter and a slag pocket to prevent slag and sand from entering the cavity. The exhaust system was designed with a total vent area of 11 cm², with vents placed at the runner end and at the highest point of the casting. This design effectively reduced gas-related casting defects such as gas holes and choke burn.
- Orthogonal experiments based on L9(3⁴) were performed with pouring temperature, filling pressure, pouring velocity, and molten metal weight as variables. The matrix analysis yielded the optimal process combination: pouring temperature 1385 °C, filling pressure 1.1 Pa, pouring velocity 0.40 m/s, and molten metal weight 155 kg. The optimized pore volume was only 0.401 cm³.
- Experimental verification on actual castings showed that the mechanical properties and metallographic structure met the technical requirements. The ultrasonic inspection did not reveal unacceptable defects. The qualification rate of the saddle castings increased from 60% to 95%, confirming the effectiveness of the proposed optimization in reducing casting defects and improving production efficiency.
In summary, this research provides a theoretical and technical basis for the “iron instead of steel” casting of tractor saddles. The combination of gating system design, numerical simulation, and orthogonal experiment optimization proves to be a powerful methodology for minimizing casting defects in large thin-wall ductile iron castings. The optimized process is now successfully applied in production, yielding high-quality saddle castings with significantly reduced defect rates.