In the field of traditional heavy industry, the production of large-scale, high-integrity sand casting parts has historically relied on costly and time-consuming trial-and-error methods. The development of a new casting process typically involves multiple iterations of physical prototyping, analysis, and modification before a sound manufacturing route is established. This approach not only consumes significant material and labor resources but also extends product development cycles considerably. With the advancement of computational power, casting simulation software has become an indispensable tool for foundries. It enables engineers to virtually analyze mold filling, solidification, and defect formation before any metal is poured. This digital prototyping significantly reduces development time and cost while improving the quality and yield of final sand casting parts. This study focuses on the application of such simulation technology to optimize the sand casting process for a critical high-pressure steam chamber, a component that operates under extreme service conditions and demands exceptional quality.
The component under investigation is a high-pressure steam chamber, a crucial part of steam turbine assemblies. This part functions as a pressurized manifold, directing and equalizing steam flow from the main stop valve to the control valves before it enters the turbine to drive the rotor. As a core pressure-retaining component, its operational environment is severe, involving continuous exposure to high temperatures (approximately 565°C) and high pressures (around 8.28 MPa). It must maintain a leak-proof performance for extended periods, often required to have a service life exceeding 20 years. These demanding conditions necessitate the use of high-alloy steel castings with superior metallurgical quality and mechanical properties. The material specified for this sand casting part is ZG15Cr2Mo1 alloy steel. The chemical composition requirements for this grade are summarized in the table below, which is critical for achieving the desired high-temperature strength and creep resistance.
| Element | Chemical Composition (wt.%) |
|---|---|
| C | ≤ 0.18 |
| Mn | 0.40 – 0.70 |
| Si | ≤ 0.60 |
| Cr | 2.00 – 2.75 |
| Mo | 0.90 – 1.20 |
| S | ≤ 0.030 |
| P | ≤ 0.030 |
The geometry of the steam chamber, after accounting for necessary machining allowances, presents significant challenges for casting. The overall envelope dimensions are 1648 mm x 620 mm x 1077 mm, with a calculated volume of 0.251 m³ and an approximate weight of 1958 kg. The wall thickness varies substantially across the component, ranging from a minimum of 30 mm to a maximum of 153 mm at certain heavy sections like the base and top bosses. This variation creates isolated thermal masses, or hot spots, which are prone to shrinkage defects during solidification. A fundamental rule in designing processes for sand casting parts is to ensure wall thicknesses remain within a suitable range for the alloy. For medium-to-large carbon and low-alloy steel sand casting parts, the recommended minimum wall thickness in sand molds is typically above 20 mm. The design of this chamber meets this criterion, but the pronounced thickness variation remains the primary challenge for achieving soundness.
The first step in process design is determining the optimal pouring position and mold parting line. For this sand casting part, two primary orientations were considered: a vertical orientation and a horizontal orientation. A horizontal pouring position was selected for several key reasons. It allows the major parting plane to be placed at the largest cross-section of the component, simplifying mold assembly and core placement. More importantly, it positions the heaviest sections of the casting (the base and top features) in the upper portion of the mold cavity. This arrangement is conducive to implementing effective feeding, as risers can be conveniently placed on these upper hot spots to compensate for solidification shrinkage. Furthermore, the horizontal orientation promotes more stable metal flow during filling compared to a deep, vertical mold, reducing the risk of turbulence and sand erosion.
Given the high pouring temperature of steel (exceeding 1550°C), the mold and core materials must exhibit high refractoriness and thermal stability. A silica sand (SiO₂ content ≥97%) bonded with phenolic resin was selected for both the mold and cores. This system provides good collapsibility and adequate strength for producing large sand casting parts. The internal cavities of the steam chamber, which include the main cylindrical chamber and several nozzle passages, require the use of dry sand cores. The core assembly was designed using three separate cores to accurately form the complex internal geometry while ensuring proper venting and stability during pouring.
Key casting process parameters were established based on standard guidelines for steel sand casting parts. The patternmaker’s contraction allowance was set at 1.8%. For the preliminary gating system design, a bottom-pouring (slide gate) system was planned to ensure non-turbulent filling. The total pouring time was calculated to achieve a sufficient metal rise speed in the mold cavity, minimizing radiant heating of the sand mold. The fundamental formula for calculating pouring time (t) is:
$$ t = \frac{G_L}{N \cdot n \cdot v_{ladle}} $$
where \( G_L \) is the total weight of metal poured (including a 7% allowance for mold dilation), \( N \) is the number of ladles, \( n \) is the number of gates per ladle, and \( v_{ladle} \) is the pouring rate of the ladle. With \( G_L \) = 2095 kg, using one ladle with a 70 mm nozzle diameter (\( v_{ladle} \) ≈ 120 kg/s), the theoretical pouring time was calculated to be approximately 17.5 seconds, rounded to 18 seconds for practical purposes. The resulting metallostatic head was checked to ensure it was sufficient to drive metal flow to the farthest parts of the cavity using the formula:
$$ h_M = L \cdot \tan(\alpha) $$
where \( h_M \) is the required minimum ferrostatic pressure head, \( L \) is the maximum flow length, and \( \alpha \) is the recommended pressure angle (taken as 6°). The designed system met this requirement.
The initial gating system featured three downsprue inlets connected to a horizontal runner bar with multiple ingates along the parting line to distribute metal evenly. No risers were included in this first design iteration. This preliminary process layout for the steel sand casting part was then modeled in ProCAST simulation software. The material properties for ZG15Cr2Mo1 were assigned, including a liquidus temperature of 1501°C. The simulation parameters were set with a pouring temperature of 1600°C, a pour time of 18 seconds, and an interfacial heat transfer coefficient of 750 W/(m²·K) between the metal and the sand mold.
The filling simulation confirmed that the gating design provided a smooth, complete fill without any visible cold shuts or misruns. However, the solidification and shrinkage analysis revealed severe defects in the as-designed process. As anticipated, large volumetric shrinkage cavities (macro-porosity) were predicted in the two massive sections at the base of the chamber and in the thick boss sections on the top. Furthermore, dispersed micro-porosity (shrinkage) was indicated in the thinner walls connecting these hot spots. These defects are unacceptable for a high-integrity pressure vessel sand casting part, as they would dramatically reduce its fatigue life and pressure containment capability under cyclic thermal and mechanical loads.
Based on the simulation results, the casting process was systematically optimized. The core strategy was to promote directional solidification towards strategically placed risers. The optimization focused on two main techniques: adding feeding risers and applying chills. First, four cylindrical risers were designed and positioned directly on the identified hot spots: two on the heavy base sections and two on the top bosses. The riser dimensions were calculated using the modulus method to ensure they remained molten longer than the sections they were intended to feed. Second, external chills were designed and placed on the mold walls adjacent to the thick sections of the base and the internal dividing walls (webs) inside the chamber. The function of these chills is to rapidly extract heat, accelerating the solidification of these regions and thereby establishing a favorable thermal gradient that directs shrinkage porosity towards the risers.
The modified process, incorporating risers and chills, was modeled again in ProCAST. The results demonstrated a marked improvement. The large shrinkage cavities were completely eliminated from the main body of the steam chamber sand casting part. The total shrinkage porosity was successfully redirected into the risers, which are later removed during machining. The final predicted shrinkage porosity in the casting itself was reduced to a minimal level of 2.71%, representing a sound casting suitable for its demanding service. The use of chills effectively controlled the solidification of the webs, although it slightly altered the shrinkage pattern in adjacent thin walls, indicating a potential area for further fine-tuning of chill size or placement.
In conclusion, this study successfully demonstrates the power of numerical simulation in designing and optimizing processes for complex, high-value sand casting parts. Starting from a basic design that would have produced defective castings, a physics-based simulation tool was used to identify critical shrinkage zones in a high-pressure steam chamber component. By interpreting these results and applying fundamental foundry engineering principles—specifically the strategic placement of risers and chills—an optimized sand casting process was developed virtually. The final simulated process showed a significant reduction in shrinkage defects, validating the design changes before any physical trial. This approach not only saves substantial cost and time but also enhances the reliability and performance of critical sand casting parts operating in extreme environments. For future work, further optimization could explore the use of insulated or exothermic riser sleeves to improve feeding efficiency, as well as different chill materials and geometries to perfect the solidification sequence in all sections of the casting.