In our investigation of lost foam castings, we have recognized that the gating system exerts a profound influence on the final quality of valve components. The lack of systematic theoretical guidance and empirical formulas for gating system design in lost foam castings has significantly constrained the advancement of this technology. To address this, we employed computer numerical simulation to observe the mold filling and solidification processes, enabling us to study how different gating parameters affect the casting quality of lost foam castings. This approach allowed us to predict defects such as shrinkage cavities, porosity, slag inclusions, gas holes, and cracks, ultimately assisting in the rational design of pouring processes for practical production.
Although valve castings have relatively simple geometries, they exhibit significant variations in wall thickness and demand high surface quality, with mandatory water pressure and air tightness tests. In this study, we took a typical valve casting produced by a factory as an example. By integrating computer simulation with orthogonal experimental design, we systematically investigated the effects of gating system type, pouring method, and in-gate cross-section shape on the quality of lost foam castings. Through analysis and comparison of the influence magnitudes of these factors, we determined the optimal gating system design scheme for lost foam castings.
Analysis Model and Determination of Factor Levels
Analysis Model
The typical valve casting we studied is made of 304 stainless steel, with a mass of 15 kg and maximum dimensions of 550 mm. The required surface roughness is Ra12.5–25.0 μm, dimensional tolerance grade 8–10, and weight tolerance grade 7–8. The geometry is relatively simple but with large differences in wall thickness, making it a challenging part for lost foam castings.
Parameter Selection and Level Determination
We used the simulation software Experto-ViewCast, developed by the Belgian National Institute of Industrial Research, which provides built-in thermophysical properties. The thermal-physical parameters of 304 stainless steel and the foam pattern are summarized in Table 1. In lost foam castings, the foam pattern consumes heat from the molten metal during thermal decomposition. Consequently, the pouring temperature for lost foam castings is typically 30–50 °C higher than that for sand casting. We adopted a dry sand negative pressure process with a pouring temperature of 1600 °C. The dry sand properties are: density 1520 kg·m−3, thermal conductivity 0.53 W·m−1·K−1, specific heat capacity 1.22 kJ·kg−1·K−1, permeability 1×10−7 cm2. The interface heat transfer coefficient between the sand mold and the molten metal or foam pattern was set to 500 W·m−2·K−1, and the negative pressure value was −0.60 MPa.
| Material | Density (kg·m−3) | Latent heat (kJ·kg−1) | Thermal conductivity (W·m−1·K−1) | Specific heat (kJ·kg−1·K−1) | Liquidus (°C) | Solidus (°C) |
|---|---|---|---|---|---|---|
| Foam material | 25 | 100 | 0.15 | 3.700 | 350 | 330 |
| 304 stainless steel | 7420 | 271.7 | 12.55 | 0.669 | 1476 | 1460 |
The orthogonal experimental method allowed us to scientifically select test conditions and arrange experiments efficiently. The general steps are: (1) clarify the experimental objective; (2) determine factors and levels; (3) select an orthogonal array; (4) design the table header; (5) formulate the test plan; (6) conduct the experiments; (7) analyze the results to identify the primary and secondary factors and the optimal level combination. The selected factors and their levels are listed in Table 2, and the corresponding schematic representations are shown in Figure 2 (note: figures are not reproduced in text to comply with guidelines).
| Level | Factor A: Gating system type | Factor B: Pouring method | Factor C: In-gate cross-section shape |
|---|---|---|---|
| 1 | Closed (F直 > F横 > F内) | Top pouring | Rectangular |
| 2 | Semi-closed (F横 > F直 > F内) | Side pouring | Circular |
| 3 | Open (F直 < F横 < F内) | Bottom pouring | Triangular |
Orthogonal Experiment Results and Analysis
We conducted nine experimental trials according to the L9 orthogonal array, as shown in Table 3. For each trial, we performed a full simulation of the mold filling process for lost foam castings using the simulation software. After computation, we examined the filling velocity distributions.
| Test No. | A: Gating system type | B: Pouring method | C: In-gate cross-section shape |
|---|---|---|---|
| 1 | 1 (Closed) | 1 (Top) | 1 (Rectangular) |
| 2 | 1 (Closed) | 2 (Side) | 2 (Circular) |
| 3 | 1 (Closed) | 3 (Bottom) | 3 (Triangular) |
| 4 | 2 (Semi-closed) | 1 (Top) | 2 (Circular) |
| 5 | 2 (Semi-closed) | 2 (Side) | 3 (Triangular) |
| 6 | 2 (Semi-closed) | 3 (Bottom) | 1 (Rectangular) |
| 7 | 3 (Open) | 1 (Top) | 3 (Triangular) |
| 8 | 3 (Open) | 2 (Side) | 1 (Rectangular) |
| 9 | 3 (Open) | 3 (Bottom) | 2 (Circular) |
According to the design principles of gating systems for lost foam castings, the velocity of molten metal entering the mold cavity should be controlled between 0.5 and 0.7 m/s to avoid surface turbulence and splashing that lead to filling defects. In our analysis, we selected 0.5 m/s as the critical velocity. For each simulation, we calculated the total duration during which the filling velocity at any location in the model exceeded or equaled 0.5 m/s. This total duration represents the cumulative time during which gases and thermal decomposition residues from the foam pattern are likely to be entrapped by the molten metal, causing filling defects. A longer total duration indicates a higher probability of forming defective lost foam castings.
The results of the orthogonal experiment are summarized in Table 4. We computed the response indicators using standard orthogonal analysis:
$$
K_{ij} = \text{sum of all test results for factor } j \text{ at level } i
$$
$$
\overline{K}_{ij} = \frac{1}{s} K_{ij}, \text{ where } s \text{ is the number of occurrences of level } i \text{ in column } j
$$
$$
R_j = \max(\overline{K}_{ij}) – \min(\overline{K}_{ij})
$$
Here, $R_j$ is the range for factor $j$. A larger range indicates that the factor has a greater influence on the experimental results, i.e., it is the more important factor.
| Test No. | Factor levels | Total duration (s) of filling velocity ≥ 0.5 m/s | ||
|---|---|---|---|---|
| A | B | C | ||
| 1 | 1 | 1 | 1 | 6.445 |
| 2 | 1 | 2 | 2 | 8.620 |
| 3 | 1 | 3 | 3 | 8.891 |
| 4 | 2 | 1 | 2 | 7.806 |
| 5 | 2 | 2 | 3 | 13.514 |
| 6 | 2 | 3 | 1 | 8.023 |
| 7 | 3 | 1 | 3 | 8.704 |
| 8 | 3 | 2 | 1 | 8.876 |
| 9 | 3 | 3 | 2 | 10.810 |
| Level averages | ||||
| $\overline{K}_{1j}$ | 7.985 | 7.652 | 7.781 | |
| $\overline{K}_{2j}$ | 9.781 | 10.337 | 9.079 | |
| $\overline{K}_{3j}$ | 8.463 | 9.241 | 10.370 | |
| Range $R_j$ | 5.387 | 8.055 | 7.765 | |
From the range values in Table 4, we see that the order of factor influence on the total duration (and thus on defect probability) is: Factor B (pouring method) > Factor C (in-gate cross-section shape) > Factor A (gating system type). The optimal level combination, based on minimizing the total duration, is A1 (closed gating system), B1 (top pouring), and C1 (rectangular in-gate cross-section).
Figure 4 (not shown) presents the trend of the average total duration for each factor level. The analysis reveals the following:
- Gating system type: The semi-closed system gave the highest average total duration (9.781 s), indicating the greatest tendency for entrainment defects. The closed system gave the lowest (7.985 s). For lost foam castings, a closed gating system ensures rapid filling of the sprue, creating a pressurized flow that minimizes aspiration of air and slag. Open and semi-closed systems cannot fill the sprue quickly, leading to higher defect risks.
- Pouring method: Side pouring and bottom pouring both resulted in longer total durations (10.337 s and 9.241 s, respectively) compared to top pouring (7.652 s). In lost foam castings, longer filling times allow more gas and residue to be trapped. Top pouring, despite the risk of mold collapse for large castings, is suitable for small-to-medium valve castings because the pouring is completed in a few seconds, before the vacuum level drops critically.
- In-gate cross-section shape: Triangular cross-section produced the highest average total duration (10.370 s), while rectangular gave the lowest (7.781 s). Triangular in-gates cause excessively high local velocities and rapid cooling, which can lead to cold shuts and misruns. Rectangular in-gates provide smoother metal flow and reduced turbulence.
Discussion
Our simulation-based orthogonal experiment clearly demonstrates that for lost foam castings of valve components, the pouring method is the most influential factor affecting casting quality, followed by the in-gate cross-section shape, and then the gating system type. The optimal design—closed gating, top pouring, and rectangular in-gate—minimizes the time during which molten metal exceeds the critical filling velocity, thereby reducing the entrapment of gas and decomposition products. This conclusion is consistent with the theoretical understanding of lost foam castings: a pressurized, quickly filled gating system combined with a smooth top-pouring arrangement yields the best quality.
It is worth noting that the absolute values of the total duration we computed depend on the specific geometry and simulation parameters. However, the relative ranking and optimal combination are robust for the typical valve casting considered. The application of computer simulation prior to actual production enables foundries to avoid costly trial-and-error experiments, saving both time and material costs while improving the quality of lost foam castings.
Conclusions
- Numerical simulation of the mold filling and solidification processes in lost foam castings allows optimization of the gating system before actual production, avoiding extensive preliminary process trials and improving casting quality.
- For lost foam castings of valve components, the pouring method has the greatest influence on casting quality, followed by the in-gate cross-section shape, while the gating system type has the smallest influence among the three factors studied.
- Under the conditions of negative pressure, the optimal gating system design for lost foam castings of valve castings is: closed gating system (F直 > F横 > F内), top pouring method, and rectangular in-gate cross-section. This combination yields the lowest probability of filling defects.