In my work on investment casting for low-carbon steel forklift heads, I encountered a significant challenge: the material exhibits high toughness and low carbon content (0.18–0.22% C), which makes it extremely difficult to break off the gating system via hammer impact. Traditional flame cutting is costly and inefficient. To solve this, I employed CAE simulation and systematically optimized the gating system design, enabling hammer knock-off while maintaining sound internal quality. This article describes my step-by-step approach, including calculation, simulation, and experimental verification, all centered on the core process of investment casting.
The forklift head is a thin-walled, hollow part with uniform wall thickness, produced by investment casting. The material must meet the mechanical properties listed in Table 1.
| ω(C) (%) | Yield Strength (MPa) | Tensile Strength (MPa) | Elongation (%) |
|---|---|---|---|
| 0.18–0.22 | 240 | 450–700 | 22 |
Because of the high elongation (22%), the material is ductile and does not fracture easily under repeated hammering. Traditional flame cutting was used for similar parts, but it raised costs and reduced efficiency. My goal was to design a gating system that allows clean breakage at the ingate neck, eliminating the need for post-cast cutting. This required careful balancing of feeding capacity and mechanical notch design.
1. Initial Casting Process Design
The 3D model of the casting is shown in the following image. The part has two main hot spots (A and B) with cross-sections of approximately 15 mm × 75 mm. I chose a single-runner, single-ingate system with a sprue cup that provides feeding through the runner. The sprue cup feeding capacity method was used to calculate the dimensions of each gating element.
1.1 Ingate Dimensions
In the sprue cup feeding capacity method, each hot spot is treated as an equivalent cylindrical body. The equivalent thermal section diameter Dc of the casting at the hot spot is related to the ingate equivalent thermal section diameter Dg by the linear relationship:
$$ D_g = k D_c $$
where Dc is the equivalent thermal section diameter of the casting hot spot (mm), and k is a weight factor obtained from a standard chart. For a rectangular section a × b, the value of Dc can be read from the well-known equivalent thermal section diameter diagram. In my case, a = 15 mm, b = 75 mm, so from the diagram I obtained Dc = 23 mm. The weight of the feeding zone assigned to one ingate was 600 g. Using the weight factor chart (inputting 600 g and the hot spot geometry), I found k = 0.93. Therefore:
$$ D_g = 0.93 \times 23 = 21.39 \, \text{mm} $$
However, due to casting geometry constraints, a circular ingate of 21.39 mm diameter could not be placed at the hot spot. Instead, I adopted a rectangular cross-section of 22 mm × 20 mm, which provides an equivalent thermal section diameter close to the required value. To facilitate hammer knock-off, I designed a sharp neck at the ingate on the side where tensile fracture is desired. The neck angle was set to 75° based on previous experience, and a 1.5 mm residual root was left below the neck to avoid excessive metal loss at the fracture. The other direction of the ingate was kept without a neck to simplify the wax die. The detailed ingate dimensions are summarized below.
| Parameter | Value |
|---|---|
| Cross-section (at hot spot) | 22 mm × 20 mm |
| Neck angle | 75° |
| Residual root height | 1.5 mm |
| Ingate height | 15 mm |
| Welding surface area (to runner) | 1950 mm² |
1.2 Runner Dimensions
In the sprue cup feeding method, the runner does not supply metal but only acts as a feeding passage. Its cross-section should be as small as possible while ensuring unobstructed flow. The required equivalent thermal section diameter Dk of the runner must satisfy:
$$ D_k \ge (1.15 \text{ to } 1.25) D_g $$
With Dg = 21.39 mm, I chose Dk = 35 mm, which is a standard size compatible with wax stick dimensions and the ingate welding area.
1.3 Sprue Cup Dimensions
The sprue cup is the primary feeding reservoir. The dimensions must provide sufficient liquid metal without excessive waste. For a square-cross-section runner, Table 2 (from the reference table of the sprue cup feeding method) gives standard combinations. With Dk = 35 mm, I selected the third group: sprue cup diameter D = 70 mm, height H = 35 mm, and other dimensions as listed. The minimum distance between ingates (L) was set to 105 mm due to part geometry.
| D (mm) | H (mm) | L (min, mm) |
|---|---|---|
| 70 | 35 | 105 |
| 80 | 40 | 120 |
| 90 | 45 | 135 |
The initial assembly tree is shown schematically in the figure (inserted above). The runner was a straight bar connecting the sprue cup to the two ingates.
2. Casting Simulation and Optimization
Before committing to production, I performed a solidification simulation using a commercial casting CAE software (anycast/Flow-3D equivalent). The mesh size was 1 mm in critical areas. Material properties were set according to the low-carbon steel composition. Only solidification analysis was conducted because the main concern was feeding.
2.1 Solidification Simulation of Initial Design
The simulation results revealed a serious issue: the ingate solidified much earlier than the hot spot, creating an “interruption” in the feeding channel. This premature solidification caused isolated liquid pools in the hot spot, leading to predicted shrinkage porosity. The shrinkage defect size was small but located exactly where subsequent machining would create critical holes. Based on my experience, such defects are unacceptable. The feeding capacity of the ingate was insufficient.
I analyzed the causes: the ingate neck cross-section was small, and the ingate itself had a relatively large surface area exposed to the shell, accelerating cooling. The feeding channel needed to be enlarged or modified to delay its solidification relative to the casting.
2.2 Optimization of the Ingate
To enhance feeding without compromising the knock-off capability, I decided to increase the ingate’s thermal mass and reduce its cooling rate. I could not simply enlarge the neck because that would make hammer fracture difficult. Instead, I focused on increasing the ingate volume and the welding surface area (the area where the ingate contacts the runner). Larger volume means more heat content, and a larger welding surface increases heat input from the runner, slowing solidification. I also reduced the ingate height from 15 mm to 10 mm by designing a sloped welding surface (see Figure 8 in the original). This slope also helped strengthen the shell during wax removal and provided a better orientation for hammering (horizontal placement of the assembly).
After several iterative CAE runs, I found that increasing the welding surface area from the original 1950 mm² to 2850 mm² eliminated the feeding interruption. The final optimized ingate geometry is shown below.
| Parameter | Initial | Optimized |
|---|---|---|
| Welding surface area (mm²) | 1950 | 2850 |
| Ingate height (mm) | 15 | 10 (sloped) |
| Cross-section at hot spot | 22×20 mm | 22×20 mm (unchanged) |
| Neck angle | 75° | 75° |
2.3 Runner and Sprue Cup Optimization
From the initial simulation, I noticed that the runner had excess liquid metal between the two ingates; the middle portion solidified last. To improve yield and reduce weight, I removed that central section. This also reduced the total metal poured. The sprue cup and remaining runner were connected via a web plate, and I added threads on the sprue cup for better shell adhesion and strength. The final assembly tree is shown in the figure (the same image as before, but now with the modified runner).
2.4 Solidification Simulation of Optimized Design
I reran the solidification simulation with the optimized gating system. The results showed no feeding interruption: the ingate solidified slightly after the hot spot, providing adequate liquid metal. Shrinkage porosity was completely eliminated in the critical areas. Only minor scattered porosity was predicted in fillet regions, which is acceptable. The simulation confirmed that the new design meets both feeding and knock-off requirements.
3. Experimental Verification
I produced 100 trial castings using the optimized process. During wax injection, the dies functioned well; no cracking or damage occurred. After drying and shell building, the shells were strong. The castings were poured at 1580°C with a pouring time of 5 seconds. After cooling, I attempted hammer knock-off by striking the sprue cup with a mallet. To my satisfaction, all castings separated cleanly at the ingate neck with minimal effort. The fracture surface was flat and free of sharp protrusions. X-ray inspection of 20 randomly selected parts showed no internal shrinkage defects in the hot spot zones. The predicted minor porosity in fillets was barely visible and did not affect subsequent machining. I then machined 95 parts (some were used for destructive testing), and all critical holes passed quality checks.
I compared the production efficiency and material yield with similar previous castings that required flame cutting. The results are shown in Table 5.
| Metric | Previous product (flame cut) | This product (hammer knock-off) | Improvement (%) |
|---|---|---|---|
| Output per shift (parts) | 260 | 600 | 130.8 |
| Process yield (%) | 55 | 68 | 23.6 |
The optimized investment casting process more than doubled the production rate and improved yield by nearly 24%, significantly reducing costs.
4. Conclusion
Through this study, I have demonstrated a successful optimization of the investment casting process for a low-carbon steel forklift head. My key findings are:
- Theoretical calculations of gating system dimensions, while useful, often deviate from real solidification behavior. CAE simulation provides rapid feedback on feeding issues and guides effective modifications. In this case, the initial ingate design caused premature solidification; simulation helped identify the root cause (insufficient thermal mass) and allowed me to adjust the ingate dimensions efficiently.
- When the ingate neck size is constrained by the requirement of hammer knock-off, feeding can still be improved by increasing the ingate volume and its contact area with the runner. A larger welding surface delays ingate solidification, ensuring adequate feed to the hot spot.
- For low-carbon, high-toughness steel castings produced by investment casting, careful design of the ingate neck (with a sharp angle and appropriate residual root) enables clean hammer fracture without compromising feeding. The combination of a sloped welding surface and reduced ingate height further enhances both feeding and breakability.
This case study illustrates how systematic use of simulation and iterative optimization can transform a challenging investment casting into a highly efficient, low-cost process. The same methodology can be applied to other similar parts.