I have been working on the production of ductile iron castings using lost foam casting for a heavy-duty reducer housing. The reducer housing is a critical component that requires high strength, toughness, wear resistance, and vibration damping. The material is QT450-10 ductile iron, and the casting weighs 112 kg. Its maximum wall thickness is 54 mm, while the minimum wall thickness is 14 mm. The geometry contains concentrated hot spots, which makes the casting prone to wrinkling and shrinkage defects. In this work, I systematically investigated the root causes and developed practical process solutions. The results show that a bottom-gating system eliminates surface wrinkling, and a novel heat-dissipation process eliminates shrinkage cavities in ductile iron castings without using risers or chills.
I selected lost foam casting because it offers excellent surface quality, high dimensional accuracy, and high process yield. However, the combination of ductile iron castings and lost foam casting introduces specific challenges. The decomposition of the foam pattern produces liquid, gaseous, and solid pyrolysis products. These products interact with the flowing metal, and if the flow is turbulent or if the pattern does not decompose uniformly, carbon-related defects such as wrinkling can appear. In addition, the geometric hot spots in the reducer housing lead to insufficient feeding during solidification, causing shrinkage porosity and shrinkage cavities. Both defects were observed at a high rate during small-batch production. I therefore focused on two objectives: first, to eliminate surface wrinkling by redesigning the gating system; second, to eliminate shrinkage defects by controlling local cooling without increasing process complexity.
| Parameter | Value |
|---|---|
| Material | QT450-10 ductile iron |
| Casting weight | 112 kg |
| Maximum wall thickness | 54 mm |
| Minimum wall thickness | 14 mm |
| Pouring temperature | 1370–1440 °C |
| Tap temperature | 1580–1600 °C |
| Vacuum level | −0.06 to −0.04 MPa |
| Vacuum holding time | 900 s |
The chemical composition of the QT450-10 ductile iron used for the reducer housing is given in Table 2. I maintained this composition throughout the trials. The carbon equivalent is within the typical range for ductile iron castings, and the magnesium and rare earth contents ensure adequate nodularity. Y-block specimens met the mechanical property requirements of QT450-10, and the nodularity was rated 2–3, which satisfies the technical requirements. However, the castings still exhibited wrinkling on the end face and shrinkage cavities at geometric hot spots.
| Element | C | Si | Mn | P | S | Mg | RE |
|---|---|---|---|---|---|---|---|
| Content (wt.%) | 3.5–4.0 | 2.0–3.0 | ≤0.45 | ≤0.05 | ≤0.025 | 0.02–0.06 | 0.015–0.04 |
Wrinkling is a common carbon-related defect in lost foam casting of ductile iron castings. It typically appears on the upper surfaces, vertical side walls, or dead corners where the metal stream arrives last or where the metal is relatively cold. The pyrolysis products from the foam pattern are complex, and their transport is influenced by many factors. When the pyrolysis products cannot escape or when the metal flow is turbulent, carbon-rich residues accumulate and form a wrinkled, orange-peel-like surface. In my original process, the gating system was top-gated, but because of the geometry of the reducer housing, the metal actually behaved as if it were a middle-to-bottom gated system. The high-temperature metal entered the top face, penetrated the thick upper wall, and then flowed downward through the thin wall in a fan-shaped pattern. After reaching the bottom, the metal rose upward. This created a turbulent filling pattern and left the top and side thick regions as cold ends, leading to severe wrinkling.
Shrinkage cavities in ductile iron castings are caused by the combined effects of liquid contraction and solidification contraction. When a hot spot cannot be fed with liquid metal during the final stages of solidification, an irregular cavity with rough walls forms. In the reducer housing, the shrinkage cavities were located in the thick sections around the bolt holes. These regions are geometric hot spots. I ruled out nodularization temperature, pouring temperature, and vacuum level as primary causes because the Y-block properties were acceptable and the casting surface showed no cold shuts or burn-on. Therefore, I concluded that the shrinkage cavities were primarily caused by the geometric hot spot, which had a high modulus and solidified later than the surrounding thin sections. The conventional solutions are to add risers or to use chills. However, for lost foam casting of ductile iron castings, risers reduce process yield and increase complexity, while chills are difficult to place and may fall off during sand filling, causing deformation and cost increases.
| Defect | Location | Main Cause | Typical Appearance |
|---|---|---|---|
| Wrinkling | End face, upper surfaces, side walls | Turbulent filling, cold ends, pyrolysis product accumulation | Orange-peel-like, rough, wrinkled surface |
| Shrinkage cavity | Thick sections around bolt holes | Geometric hot spot, insufficient feeding | Irregular internal cavity with rough walls |
To solve the wrinkling problem, I redesigned the gating system from a top-gated arrangement to a center closed bottom-gated arrangement. The goal was to achieve a purely bottom-gated filling pattern so that the high-temperature metal rises steadily from the bottom to the top. In this way, the cold metal and the incompletely vaporized foam products are pushed to the top machining allowance, where they can be removed. I also removed the upper slag traps. This change eliminated the turbulent middle-to-bottom filling behavior and produced a smooth, laminar-like filling front. The theoretical calculations for the new gating system are summarized below.
The pouring time was calculated using the following relationship:
$$ t = 24 S $$
where \( t \) is the pouring time in seconds and \( S \) is a shape factor related to the casting wall thickness. For the bottom-gated system, the average static pressure head was calculated as:
$$ H_p = 34 \text{ cm} $$
The minimum cross-sectional area of the ingate was calculated using the standard equation for bottom gating:
$$ A_{\min} = 3.46 \text{ cm}^2 $$
The above calculation is based on sand casting. However, in lost foam casting of ductile iron castings, the ingate area is usually slightly larger than that used in sand casting because the foam pattern must decompose and the flow resistance is different. Based on experience, the ingate area for lost foam casting is often selected between 3.5 and 12 cm². For the reducer housing, I designed a center gating system with four ingates. Each ingate had a cross-section of \((7 + 1 \times 40)\) mm, giving a total ingate area of approximately \(11.2\) to \(12.8\) cm². The sprue height was set to 480 mm, and the pressure head was 200 mm. Table 3 compares the original and optimized gating systems.
| Feature | Original Top-Gated System | Optimized Bottom-Gated System |
|---|---|---|
| Gating type | Top gate, but actual flow was middle-to-bottom | Center closed bottom gate |
| Filling pattern | Turbulent, fan-shaped downward flow, then upward | Steady upward filling from bottom to top |
| Cold end location | Top and side thick regions | Top machining allowance |
| Slag trap | Upper slag trap present | Upper slag trap removed |
| Wrinkling defect | Severe, batch occurrence | Eliminated |
After the gating system was changed to bottom gating, I conducted trial production and then a batch production of 2000 pieces. The surface quality was satisfactory, and no batch wrinkling occurred. The bottom-gated system allowed the metal to fill the cavity smoothly, and the top machining allowance acted as a collector for the cold metal and pyrolysis products. This is a simple and effective solution for wrinkling in lost foam casting of ductile iron castings.
To solve the shrinkage cavity problem, I developed a new heat-dissipation process. The core idea is to change the local cooling structure by increasing the heat-dissipation surface area, thereby lowering the local modulus at the hot spot. During solidification, the negative pressure draws cold air through the sand and the heat-dissipation fins, which removes a large amount of heat. This creates a chilling effect similar to that of a chill, but without the handling difficulties of placing metallic chills. The fins are made of foam and are attached to the hot spot before coating, drying, and sand filling. Because the fins are part of the foam pattern, they decompose during pouring and leave no residue. The local mold modulus is reduced, and the solidification sequence becomes closer to directional solidification. The heat-dissipation process is simple, has little effect on the casting process, and almost does not reduce the process yield of ductile iron castings.
The modulus of a casting section is defined as:
$$ M = \frac{V}{A_s} $$
where \( M \) is the modulus, \( V \) is the volume, and \( A_s \) is the heat-dissipation surface area. By attaching heat-dissipation fins to the hot spot, the effective surface area \( A_s \) increases, so the local modulus \( M \) decreases. The solidification time is proportional to the square of the modulus:
$$ t_s = K M^2 $$
where \( t_s \) is the solidification time and \( K \) is a constant. A lower modulus means a shorter solidification time, so the hot spot solidifies earlier and can be fed by the surrounding liquid metal. In this way, shrinkage cavities are avoided. The negative pressure in lost foam casting continuously draws air through the sand, and the fins create micro-channels for air flow. The air exchanges heat with the fins and the casting, producing a local quench. This is the essential mechanism of the heat-dissipation process.
| Parameter | Value |
|---|---|
| Fin material | Foam sheet |
| Fin dimensions | 50 mm × 30 mm × 7 mm |
| Number of fins | 12 |
| Location | Geometric hot spots around bolt holes |
| Attachment stage | Pattern assembly and gluing |
| Process steps after attachment | Coating, drying, sand filling, pouring |
I applied the heat-dissipation fins to the reducer housing in the regions where the shrinkage cavities had been observed. The original process produced shrinkage cavities in the bolt holes, as shown in the previous trials. The hot spot was a thick section that could not be fed adequately. After attaching 12 fins, the local modulus was reduced, and the cooling rate increased. I then performed machining verification. The bolt holes were sound, and no shrinkage defects were found. A batch of 2000 pieces was produced continuously, and no batch shrinkage cavities reappeared after machining. This confirms that the heat-dissipation process is effective for ductile iron castings produced by lost foam casting.
| Method | Principle | Advantages | Disadvantages |
|---|---|---|---|
| Riser | Provides liquid metal to compensate contraction | Effective for large hot spots | Reduces process yield, increases complexity |
| Chill | Increases local cooling rate | Promotes directional solidification | Difficult to place in lost foam, may fall off, causes deformation |
| Heat-dissipation fin | Increases surface area and uses negative pressure air flow to remove heat | Simple, low cost, no yield loss, no chill placement problems | Requires careful placement and fin design |
The heat-dissipation process is especially suitable for lost foam casting because the foam fins are integrated into the pattern. There is no need to insert metallic chills into the sand, which avoids chill displacement and casting distortion. The fins also do not interfere with the coating or sand filling. During pouring, the fins decompose, and the negative pressure draws air through the resulting channels. The air removes heat from the hot spot. The local cooling rate increases, and the hot spot solidifies before the surrounding thin sections. This is the opposite of the usual problem, where the hot spot solidifies last. By reversing the solidification sequence, shrinkage cavities are eliminated. The process yield remains high because no riser is added. The reducer housing casting weight is unchanged, and no extra metal is wasted.
I also considered the effect of the heat-dissipation fins on the surface quality of the ductile iron castings. The fins are attached to the pattern surface, so they leave shallow impressions on the casting surface after decomposition. However, these impressions are located on the machining allowance or in non-critical areas. In the case of the bolt holes, the impressions are inside the holes and are removed during machining. The surface quality of the final casting is not affected. The fins are thin and small, so they do not change the overall dimensions of the casting. The dimensional accuracy of the ductile iron castings is maintained.
| Process Stage | Original Process | Optimized Process |
|---|---|---|
| Gating system | Top gate with upper slag trap | Center closed bottom gate, no upper slag trap |
| Filling behavior | Turbulent, cold ends at top and sides | Smooth upward filling, cold metal collected at top allowance |
| Hot spot control | None | Heat-dissipation fins attached to hot spots |
| Wrinkling defect | Batch occurrence | Eliminated |
| Shrinkage cavity | Batch occurrence in bolt holes | Eliminated |
| Process yield | Reduced by risers if used | Maintained high |
| Manufacturing complexity | High due to gluing and turbulent filling | Low, simple fin attachment |
The experimental validation was performed on a production scale. I produced 2000 reducer housings using the optimized process. The results are summarized in Table 7. The surface wrinkling defect was completely eliminated. The shrinkage cavities in the bolt holes were eliminated. The mechanical properties of the Y-block specimens remained within the QT450-10 specification. The nodularity was still 2–3. The process yield was high, and the production efficiency was improved because the gating system was simpler and the gluing process was less complicated. The heat-dissipation fins added a small amount of manual work, but this was offset by the elimination of rework and scrap caused by shrinkage defects.
| Validation Item | Result |
|---|---|
| Batch size | 2000 pieces |
| Surface wrinkling | None observed |
| Shrinkage cavity in bolt holes | None observed after machining |
| Y-block mechanical properties | Meet QT450-10 |
| Nodularity | 2–3 |
| Process yield | High |
| Production stability | Stable, no batch defects |
The results demonstrate that the two defects have different physical origins and require different solutions. Wrinkling is a filling-related defect caused by turbulent flow and incomplete decomposition of the foam pattern. Shrinkage is a solidification-related defect caused by geometric hot spots. By separating these two problems and addressing them independently, I achieved a robust process for ductile iron castings. The bottom-gating system solves the filling problem, and the heat-dissipation fins solve the solidification problem. The combination of these two solutions is greater than the sum of their parts because the bottom-gating system also reduces turbulence and ensures that the heat-dissipation fins do not disturb the filling front.
The heat-dissipation process can be further optimized by adjusting the fin dimensions and the number of fins. The fin surface area determines the local cooling capacity. The negative pressure level also affects the air flow rate and the heat transfer coefficient. In my experiments, a fin size of 50 mm × 30 mm × 7 mm and a quantity of 12 was sufficient for the reducer housing. For other ductile iron castings with different hot spot geometries, the fin design should be customized. The fin thickness should be large enough to provide a flow channel but small enough to decompose quickly. The fin attachment must be secure so that the fins do not fall off during coating or sand filling. I used a standard foam glue and allowed sufficient drying time.
| Variable | Effect on Cooling | Recommended Practice |
|---|---|---|
| Fin surface area | Larger area increases heat removal | Size fins to match hot spot modulus |
| Fin thickness | Thicker fins provide larger channels but decompose more slowly | Use 5–10 mm thickness |
| Number of fins | More fins increase cooling but may affect surface finish | Place only at hot spots |
| Vacuum level | Higher vacuum increases air flow and cooling | Maintain −0.06 to −0.04 MPa |
| Fin location | Must cover the hot spot completely | Center fins on the hot spot |
The chemical composition and nodularity of the ductile iron castings were not changed by the new process. The carbon, silicon, manganese, phosphorus, sulfur, magnesium, and rare earth contents remained within the specified ranges. The spheroidization and inoculation treatments were unchanged. Therefore, the mechanical properties were maintained. The heat-dissipation fins do not introduce any alloying elements or impurities because they are made of the same foam material as the pattern. They decompose into gaseous products that are drawn away by the vacuum system. No carbon pickup or gas porosity was observed in the castings. The surface quality and internal soundness were both improved.
The economic benefits of the new process are significant. The original process required expensive rework and scrapping of defective ductile iron castings. The new process eliminates wrinkling and shrinkage cavities, so the scrap rate is greatly reduced. The bottom-gating system is simpler than the original top-gating system because the upper slag trap is removed. The heat-dissipation fins are inexpensive and easy to attach. The process yield is high because no risers are used. The overall production cost is lower. The process is also more environmentally friendly because less metal is wasted and less energy is consumed in rework.
| Economic Factor | Original Process | Optimized Process |
|---|---|---|
| Scrap rate due to wrinkling | High | Near zero |
| Scrap rate due to shrinkage | High | Near zero |
| Riser cost | High if used | None |
| Chill cost | High if used | None |
| Fin material cost | None | Very low |
| Labor for gluing | High due to complex gating | Low, simple fin attachment |
| Process yield | Reduced | High |
In summary, I have developed a comprehensive solution for wrinkling and shrinkage in lost foam casting of ductile iron castings. The key findings are as follows. First, the original top-gated system caused a middle-to-bottom filling pattern with turbulent flow, which produced wrinkling on the top and side surfaces. By changing to a center closed bottom-gated system, the filling became smooth and upward, and the cold metal and pyrolysis products were collected in the top machining allowance. This eliminated wrinkling. Second, the geometric hot spots in the thick sections around the bolt holes caused shrinkage cavities because they could not be fed. By attaching foam heat-dissipation fins to these hot spots, I increased the local heat-dissipation surface area, reduced the local modulus, and used the negative pressure air flow to create a chilling effect. This promoted directional solidification and eliminated shrinkage cavities. Third, the new process was validated in a batch of 2000 pieces. No wrinkling or shrinkage defects were found. The mechanical properties and nodularity met the requirements. The process is simple, robust, and cost-effective. It can be applied to other ductile iron castings produced by lost foam casting.
The heat-dissipation process is a new method for solving shrinkage defects in lost foam casting. It differs from traditional risers and chills because it does not add metal or metallic inserts. Instead, it uses the foam pattern itself to create cooling channels. The negative pressure in lost foam casting is essential because it drives air through the channels. Without negative pressure, the cooling effect would be much weaker. The fin design must consider the balance between cooling capacity and pattern integrity. If the fins are too large, they may weaken the pattern or leave large impressions. If they are too small, they may not provide sufficient cooling. I recommend using fins with a thickness of 5–10 mm and a surface area of 1000–2000 mm² per fin, depending on the hot spot size. The number of fins should be sufficient to cover the hot spot area. In my case, 12 fins of 50 mm × 30 mm × 7 mm were used.
| Design Parameter | Recommended Range | Optimal for Reducer Housing |
|---|---|---|
| Fin thickness | 5–10 mm | 7 mm |
| Fin length | 30–80 mm | 50 mm |
| Fin width | 20–50 mm | 30 mm |
| Number of fins | 4–20 | 12 |
| Fin surface area per fin | 1000–2000 mm² | 1500 mm² |
| Attachment method | Foam glue | Standard foam glue |
The bottom-gating system also has specific design requirements. The pouring time must be long enough to allow the foam pattern to decompose completely but short enough to avoid cold shuts. The relationship \( t = 24 S \) provided a good starting point. The average static pressure head \( H_p = 34 \text{ cm} \) ensured that the metal filled the cavity steadily. The minimum ingate area \( A_{\min} = 3.46 \text{ cm}^2 \) was calculated for sand casting, but I increased it to \( 11.2–12.8 \text{ cm}^2 \) for lost foam casting. The sprue height was 480 mm, and the pressure head was 200 mm. These values can be adjusted for other ductile iron castings. The key is to avoid turbulence and to maintain a smooth upward filling front. The ingates should be distributed symmetrically around the center to avoid localized hot spots. In my case, four ingates were used.
The heat-dissipation fins are attached after the pattern is assembled. The pattern is then coated with refractory coating and dried. The fins are coated along with the pattern, so they are protected from direct contact with the sand. During pouring, the metal enters the cavity and the foam pattern decomposes. The fins decompose as well, and the negative pressure draws air through the channels left by the fins. The air removes heat from the metal and the sand. The local cooling rate increases. The hot spot solidifies earlier. The surrounding liquid metal can then feed the hot spot because the hot spot is no longer the last region to solidify. This is the essence of the heat-dissipation process. It is a dynamic process that depends on the negative pressure and the air flow. The vacuum system must be maintained throughout pouring and solidification. In my experiments, the vacuum holding time was 900 s, which was sufficient for the reducer housing.
| Step | Action | Purpose |
|---|---|---|
| 1 | Assemble foam pattern | Prepare the shape of the ductile iron casting |
| 2 | Attach heat-dissipation fins at hot spots | Increase local surface area |
| 3 | Apply refractory coating | Protect pattern and fins |
| 4 | Dry coating | Remove moisture |
| 5 | Place pattern in sand flask | Support pattern |
| 6 | Fill sand and compact | Create mold |
| 7 | Apply vacuum | Draw air through sand and fins |
| 8 | Pour molten ductile iron | Fill cavity |
| 9 | Maintain vacuum during solidification | Continue heat removal |
| 10 | Release vacuum and shake out | Remove casting |
The process has been proven in production. I have produced 2000 reducer housings with no batch defects. The surface wrinkling is gone, and the shrinkage cavities in the bolt holes are gone. The mechanical properties are acceptable. The process yield is high. The production cost is lower. The process is simple and easy to implement. I believe this method can be applied to other ductile iron castings with geometric hot spots. It is especially useful for lost foam casting, where traditional chills are difficult to use. The heat-dissipation fins are a practical and effective solution. They use the inherent negative pressure of the lost foam process to create a local chilling effect. This is a new approach to solving shrinkage defects in ductile iron castings.
Future work could include the optimization of the fin geometry using computer simulation. The heat transfer and solidification can be modeled to predict the optimal fin size and placement. The effect of different vacuum levels on the cooling rate can also be studied. The heat-dissipation process could be combined with other feeding methods for very large hot spots. However, for the reducer housing, the current design is sufficient. The process has been validated and is ready for production. I am confident that this method will help other foundries produce high-quality ductile iron castings with lost foam casting.
| Conclusion | Finding |
|---|---|
| Wrinkling cause | Turbulent filling and cold ends due to top gating |
| Wrinkling solution | Center closed bottom gating with top machining allowance |
| Shrinkage cause | Geometric hot spots with insufficient feeding |
| Shrinkage solution | Heat-dissipation fins with negative pressure air flow |
| Validation | 2000 pieces, no batch defects |
| Benefits | Simple, high yield, low cost, robust |
| Application | Ductile iron castings in lost foam casting |
In conclusion, I have solved the wrinkling and shrinkage problems in ductile iron castings produced by lost foam casting. The bottom-gating system eliminated wrinkling by providing smooth upward filling. The heat-dissipation fins eliminated shrinkage by increasing local cooling and reducing the hot spot modulus. The process was validated in a batch of 2000 reducer housings. The results show that the new process is effective, simple, and economical. It can be applied to other ductile iron castings with similar defects. The heat-dissipation process is a new contribution to the field of lost foam casting. It uses the negative pressure of the process to create a local chilling effect without metallic chills or risers. This is a significant improvement for the production of high-quality ductile iron castings.