In the modern pulp and paper industry, the pulping rotor is a critical component that directly influences the efficiency of fiber separation and the overall energy consumption of the process. Traditional manufacturing of such rotors typically relies on wooden patterns and conventional sand casting. However, the complex geometry—featuring eight streamlined blades, a hollow inner cavity, and stringent dimensional tolerances—makes wooden pattern fabrication both time-consuming and expensive. The pattern-making cycle alone can exceed 20 days, and the cost can reach tens of thousands of RMB. Moreover, the inherent limitations of wooden patterns, such as the need for draft angles and the risk of dimensional inaccuracies, often lead to defects like shrinkage porosity, gas holes, and slag inclusions. These defects are classic examples of a sand casting defect that degrades mechanical properties and service life.
To overcome these challenges, we adopted an advanced approach combining 3D printed sand molds with gravity casting for the production of a ZG06Cr13Ni4Mo martensitic stainless steel pulping rotor. 3D printing eliminates the need for physical patterns by directly fabricating sand molds from a digital model, reducing the mold-making lead time to just one day and cutting costs significantly. Furthermore, 3D printed sand molds allow for complex internal cavities and zero draft angles, which is particularly beneficial for the intricate blade geometry of the rotor. We also employed Procast simulation software to optimize the casting design, particularly the riser system, to minimize the risk of shrinkage and other sand casting defect types. This paper presents a comprehensive study of the entire process, detailing the material requirements, structural features, casting process design, simulation, and final quality validation.
Material Requirements and Structural Characteristics
The pulping rotor is specified to be made of ZG06Cr13Ni4Mo, a medium-to-high-strength martensitic stainless steel known for its excellent corrosion resistance, wear resistance, and good mechanical properties. The chemical composition and mechanical property requirements are listed in Tables 1 and 2, respectively. Additionally, the rotor must pass ultrasonic inspection in accordance with ASTM A609-12 grade 2, with no internal defects exceeding the allowable limits.
| C | Si | Mn | S | P | Cr | Ni | Mo |
|---|---|---|---|---|---|---|---|
| ≤0.06 | ≤0.80 | ≤1.00 | ≤0.035 | ≤0.025 | 11.5–13.5 | 3.5–5.0 | 0.4–1.0 |
| Yield Strength \(R_{p0.2}\) (MPa) | Tensile Strength \(R_m\) (MPa) | Elongation \(A_5\) (%) | Reduction of Area \(Z\) (%) | Impact Energy \(KV\) (J) | Hardness (HBW) |
|---|---|---|---|---|---|
| ≥550 | ≥750 | ≥15 | ≥35 | ≥50 | 221–294 |
The rotor has an outer diameter of 945 mm, a height of 158 mm, and a final weight of 156 kg. Its structure consists of a central disc surrounded by eight streamlined blades. To reduce weight and improve dynamic balance, the rotor is designed with a hollow inner cavity. The blade profile is critical for the hydraulic performance, and any dimensional deviation directly affects the rotor’s balancing and operational stability. Traditional wooden pattern casting often introduces sand casting defect issues such as misruns or shrinkage at the blade tips, leading to rejection. Therefore, achieving high dimensional accuracy and sound internal quality was our primary objective.
Casting Process Design
Shrinkage Allowance
For ZG06Cr13Ni4Mo stainless steel, the linear shrinkage during solidification is approximately 2.0%. We applied this shrinkage factor to all dimensions of the 3D model before generating the sand mold geometry.
Gating System
A bottom-gating system was adopted to ensure smooth filling and minimize turbulence. The ingate diameter was 60 mm. The molten steel enters the mold cavity from the bottom and rises steadily, allowing gases and light inclusions to float upward into the risers. This design significantly reduces the occurrence of gas-related sand casting defect.
Machining Allowance
We assigned machining allowances as follows:
- Top surface: 8 mm
- Outer diameter: 10 mm
- Bottom surface: 6 mm
- Central shaft hole: not cast, left for machining
The resulting total weight of the raw casting (including risers and gating) was 187 kg, with a steel yield (casting weight / poured weight) of 60% after optimization.
Riser Design
The primary riser was a cylindrical blind riser with diameter 200 mm and height 280 mm, located at the center of the rotor. Its modulus was calculated as:
$$ M_{\text{riser}} = \frac{V}{A} = \frac{\pi (0.1)^2 (0.28)}{2\pi (0.1)(0.28) + 2\pi (0.1)^2} \approx 0.045 \, \text{m} = 4.5 \, \text{cm} $$
This ensures the riser solidifies last and feeds the heavy central section. Additionally, eight small insulated risers (diameter 70 mm, height 100 mm) were placed at the root of each blade to prevent shrinkage at these critical junctions. Their effective modulus was 1.8 cm. Finally, each blade tip had a small vent riser (diameter 20 mm) to release gas and indicate fill completion. The total poured steel weight was 312 kg.
3D Printed Sand Mold Design
The sand molds were printed using a VX2000 3D printer, which has a maximum build volume of 2000 mm × 1000 mm × 1000 mm, a dimensional accuracy of ≤0.3 mm, and a layer thickness of 0.3 mm. We used GS15-grade silica sand with 1.2% 3DP resin binder and 0.3% 3DP curing agent. The printed sand strength reached 1.4 MPa. The 3D printed cope and drag are illustrated conceptually—note that in our work, the digital model was directly sliced and printed without any physical pattern, which eliminated the need for draft angles and greatly reduced the risk of a sand casting defect caused by mismatched cores or part lines.
Melting and Pouring
We used a duplex melting process with an alkaline electric arc furnace and an AOD (argon‑oxygen decarburization) refining furnace. To ensure the required corrosion and wear resistance, the carbon content was controlled below 0.03%, and sulfur and phosphorus were strictly limited. After tapping, the molten steel was held in the ladle for more than 5 minutes to allow oxides and slag to float. The pouring temperature was 1560–1580 °C, and a bottom-pour ladle with a 50 mm nozzle was used.
Heat Treatment
The rotor underwent normalizing and tempering. Normalizing was performed at 1020±10 °C for 2 hours, followed by tempering at 600±10 °C for 4 hours. The heating rate was controlled: ≤100 °C/h between 300–600 °C, and ≤80 °C/h between 600–1020 °C. This heat treatment achieves the required martensitic structure with good toughness and hardness.
Simulation of the Casting Process
We used Procast software to model the filling and solidification. The mesh was generated with the following boundary conditions:
- Material: ZG06Cr13Ni4Mo
- Heat transfer coefficient at the mold-metal interface: 500 W/(m²·K)
- Pouring temperature: 1580 °C
- Initial ambient temperature: 25 °C
- Pouring time: 6 seconds
- Pouring rate: 50 kg/s
- Cooling: still air
Figure 7 in our simulation showed the temperature distribution during solidification; the last solidifying regions were the central riser and the blade roots. Figure 8 predicted potential shrinkage porosity. Based on the simulation, we adjusted the size of the small risers at the blade roots to ensure adequate feeding. The final design eliminated any sand casting defect caused by shrinkage, as confirmed by the subsequent ultrasonic inspection.
Results and Validation
After 3D printing, core assembly, pouring, shakeout, cutting of gates and risers, heat treatment, and shot blasting, we performed a series of quality checks on the rotor.
Dimensional Inspection
All eight blades were measured using a surface plate and height gauges. The height of each blade was uniform and within the tolerance of ±0.5 mm, demonstrating the high accuracy of 3D printed sand molds. The dimensional deviation from the CAD model was less than 0.3 mm, which is a significant improvement over traditional wooden pattern casting, where draft angles and pattern wear often cause dimensional scatter and increase the probability of a sand casting defect related to geometry.
Chemical Composition and Mechanical Properties
Tensile test specimens were machined from the casting. The results are summarized in Table 3, all values meeting the specification. The ultrasonic inspection performed according to ASTM A609-12 grade 2 revealed no indications of unacceptable internal defects. The absence of shrinkage, gas pores, and inclusions confirms that our optimized risering and gating system, along with the clean 3D printed sand, effectively mitigated common sand casting defect issues.
| Property | Measured Value | Requirement |
|---|---|---|
| Yield Strength \(R_{p0.2}\) (MPa) | 682 | ≥550 |
| Tensile Strength \(R_m\) (MPa) | 832 | ≥750 |
| Elongation \(A_5\) (%) | 18.5 | ≥15 |
| Reduction of Area \(Z\) (%) | 42 | ≥35 |
| Impact Energy \(KV\) (J) | 68 | ≥50 |
| Hardness (HBW) | 255 | 221–294 |
Cost and Cycle Time Comparison
The 3D printed sand mold volume was 0.48 m³, with a material cost of 0.35 ten thousand RMB per cubic meter. Thus, the total sand cost was only 0.168 ten thousand RMB. In contrast, a wooden pattern for the same rotor would cost approximately 3 ten thousand RMB. Moreover, the pattern fabrication cycle of 20+ days was reduced to a single day for 3D printing. For low-volume production of complex rotors, 3D printed sand molds are clearly more economical and rapid, while also eliminating the pattern-induced sand casting defect risk.
Conclusion
This study demonstrates a successful application of 3D printed sand molds combined with gravity casting for the production of a complex pulping rotor made of ZG06Cr13Ni4Mo stainless steel. The following conclusions can be drawn:
- Procast simulation effectively guided the riser design, achieving a casting yield of 60% and eliminating shrinkage porosity and other sand casting defect types.
- 3D printed sand molds shortened the manufacturing lead time from 20+ days to 1 day for the mold, and reduced tooling costs by nearly 95% compared to wooden patterns.
- The high dimensional accuracy of 3D printed sand (≤0.3 mm) eliminated the need for draft angles and minimized the occurrence of sand casting defect related to misalignment or core shift. The final rotor met all chemical, mechanical, and ultrasonic inspection requirements.
The combination of additive manufacturing and simulation offers a robust and cost-effective route for producing high-quality steel castings with complex geometries, especially when traditional pattern-based methods would be prohibitively expensive or time-consuming. The approach can be extended to other similar components in the pulp and paper industry and beyond.