In the field of power generation and heavy machinery, steel castings play a pivotal role due to their superior mechanical properties and ability to withstand extreme operational conditions. As a critical component in steam turbines, the high-pressure steam chamber operates under high temperatures and pressures, necessitating impeccable integrity and longevity. This study focuses on the sand casting process for a high-pressure steam chamber made of ZG15Cr2Mo1 alloy steel, employing numerical simulation to optimize the casting design and mitigate defects. The use of simulation software like ProCAST has revolutionized the steel casting industry by enabling virtual trial-and-error, reducing development time and costs significantly. Herein, we present a comprehensive analysis of the casting process, from initial design based on theoretical calculations to iterative improvements guided by simulation results, all aimed at enhancing the quality and reliability of steel castings for such demanding applications.
The high-pressure steam chamber, as part of a steam turbine, is subjected to continuous operation at temperatures around 565°C and pressures up to 8.28 MPa. These conditions mandate that the steel casting exhibits no leaks or failures over a lifespan of 20 years. The material of choice, ZG15Cr2Mo1, is a low-alloy heat-resistant steel casting with specific chemical composition requirements to ensure high-temperature strength and corrosion resistance. The chemical composition of ZG15Cr2Mo1 steel casting is summarized in Table 1.
| C | Mn | Si | Cr | Mo | S | P |
|---|---|---|---|---|---|---|
| ≤0.18 | 0.40–0.70 | ≤0.60 | 2.00–2.75 | 0.90–1.20 | ≤0.030 | ≤0.030 |
Note: For every 0.01% reduction in C content above the upper limit, Mn content may increase by 0.04%, but Mn shall not exceed 1.2%.
The casting geometry, after accounting for machining allowances, has overall dimensions of 1648 mm × 620 mm × 1077 mm, with a volume of 0.251 m³ and a mass of approximately 1957.8 kg. The wall thickness varies from a minimum of 30 mm to a maximum of 153 mm, which aligns with the recommended thickness range for steel castings in sand casting processes. A thorough wall thickness analysis confirms that the design meets the minimum and critical thickness criteria, preventing issues like premature solidification or excessive shrinkage.
In steel casting, the design of the casting process is paramount to avoid defects such as shrinkage cavities, porosity, and misruns. We began by determining the optimal pouring position and parting line. Two potential configurations were considered: vertical and horizontal pouring. After evaluation, the horizontal pouring position was selected for several reasons. Firstly, it allows the parting line to coincide with the largest cross-section, simplifying mold assembly and core placement. Secondly, it facilitates the placement of risers on the thicker sections, which are positioned at the top due to the orientation, promoting directional solidification. Lastly, this configuration avoids the use of chaplets, streamlining the process. The parting line is set at the base of the steam chamber, ensuring easy mold separation and reducing the risk of mismatches.
The mold and core design for this steel casting involves three sand cores made from phenolic resin-bonded sand, which offers high refractoriness to withstand the elevated pouring temperatures of steel. The cores are strategically placed to form the internal cavities of the steam chamber, including the main chamber and the inlet/outlet ports. The mold material uses silica sand with SiO₂ content exceeding 97%, coated with alumina-based alcohol paint to enhance surface finish and prevent metal penetration. The design ensures proper venting and gating to achieve a smooth fill.
Casting process parameters were established based on industry standards for steel castings. Given the small-batch production, manual molding with self-setting resin sand was chosen. The dimensional tolerance grade is CT13 per GB/T 6414-2017, and the weight tolerance grade is MT12 per GB/T 11351-2017, with a weight tolerance of 8%. The constrained linear shrinkage rate is set at 1.8% to account for the contraction of steel during solidification. The pouring temperature is critical in steel casting; for ZG15Cr2Mo1, the liquidus temperature is 1501°C, and we set the pouring temperature at 1600°C to ensure adequate fluidity while minimizing thermal shock.
The gating system design is based on theoretical calculations to achieve a controlled fill. For medium-to-large steel castings like this, an open gating system with bottom pouring is preferred. The pouring time is calculated using the formula:
$$ t = \frac{G_L}{N \cdot n \cdot v_{\text{pack}}} $$
where \( t \) is the pouring time in seconds, \( G_L \) is the total weight of molten steel in the mold (including allowances for mold expansion and gating/riser systems), \( N \) is the number of ladles, \( n \) is the number of holes per ladle, and \( v_{\text{pack}} \) is the pouring speed per hole. With \( G_L = 2094.846 \, \text{kg} \) (including a 7% mold expansion allowance), \( N = 1 \), \( n = 1 \), and \( v_{\text{pack}} = 120 \, \text{kg/s} \) for a 70 mm diameter nozzle, the calculated pouring time is 17.46 s, rounded to 18 s for practicality.
To verify the adequacy of the metal rise speed, we use:
$$ v_L = \frac{h_C}{t} $$
where \( v_L \) is the rise speed in mm/s, and \( h_C = 620 \, \text{mm} \) is the height of the casting in the pouring position. This yields \( v_L = 35.5 \, \text{mm/s} \), which exceeds the minimum required 35 mm/s for complex steel castings, ensuring a non-turbulent fill. The gating ratio for an open system is typically:
$$ \Sigma A_{\text{pack}} : \Sigma A_{\text{sprue}} : \Sigma A_{\text{runner}} : \Sigma A_{\text{ingate}} = 1.0 : (1.8 \text{ to } 2.0) : (1.8 \text{ to } 2.0) : (2.0 \text{ to } 2.5) $$
We designed three evenly distributed ingates at the parting plane to promote uniform temperature distribution. The minimal remaining head pressure is checked with:
$$ h_M = L \tan \alpha $$
where \( h_M \) is the minimal residual head in mm, \( L = 790 \, \text{mm} \) is the horizontal distance from the sprue to the farthest point, and \( \alpha = 6^\circ \) is the pressure angle. With \( h_M = 83 \, \text{mm} \), the condition \( h_M \geq L \tan \alpha \) is satisfied, ensuring sufficient pressure to feed the casting.
Initial simulations using ProCAST software were conducted to analyze the filling and solidification behavior of this steel casting. The mesh model included the casting, gating system, and mold, with parameters as listed in Table 2.
| Parameter | Value |
|---|---|
| Pouring Temperature | 1600°C |
| Pouring Time | 18 s |
| Heat Transfer Coefficient (Metal-Mold) | 750 W/(m²·K) |
| Liquidus Temperature | 1501°C |
| Solidus Temperature | Approx. 1450°C (estimated) |
| Mold Material | Silica Sand |
| Ambient Temperature | 25°C |
The filling simulation showed a smooth fill without any short runs or cold shuts, validating the gating design. However, the solidification simulation revealed significant shrinkage defects in the steel casting. As depicted in the defect prediction results, large shrinkage cavities were observed in the thick sections of the steam chamber base and the top bosses, along with dispersed microporosity in other areas. These defects arise due to the formation of hot spots where solidification is delayed, creating isolated liquid pools that shrink without adequate feeding. The Niyama criterion, often used to predict shrinkage porosity in steel castings, can be expressed as:
$$ G / \sqrt{\dot{T}} \leq C $$
where \( G \) is the temperature gradient, \( \dot{T} \) is the cooling rate, and \( C \) is a material constant. Low values of this ratio indicate a high risk of porosity, which aligns with the simulated defect locations.
To address these issues, we modified the casting design by adding risers and chills. Riser design was based on the modulus method, where the riser modulus \( M_r \) must exceed the casting modulus \( M_c \) at the hot spots. The modulus is calculated as volume divided by surface area. For the thick base sections, \( M_c \) was approximately 0.05 m, leading to the selection of cylindrical risers with dimensions derived from standard charts. Four risers were placed atop the thick regions, as shown in the optimized design. Additionally, external chills made of cast iron were inserted near the thin sections adjacent to the thick areas, such as the inner walls of the steam chamber base and the partition plates between inlet ports. Chills accelerate cooling in these zones, promoting directional solidification towards the risers.
The improved design was re-simulated in ProCAST, and the results demonstrated a marked reduction in defects. The shrinkage cavities were successfully relocated to the risers, with no major cavities remaining in the steel casting itself. The total shrinkage porosity was reduced to 2.71%, as quantified by the software’s defect analysis module. However, some microporosity persisted in the top boss regions due to the chilling effect altering the solidification pattern. This highlights the complex interplay between cooling rates and feeding in steel casting processes.
Further optimization could involve adjusting riser sizes, using insulating sleeves on risers to enhance feeding efficiency, or modifying chill dimensions. The simulation approach allows for rapid iteration without physical trials. For instance, we can explore the impact of varying pouring temperatures or gating designs on defect formation. The relationship between pouring temperature \( T_p \) and defect severity can be modeled empirically. A higher \( T_p \) may improve fluidity but increase shrinkage, while a lower \( T_p \) might lead to mistruns. An optimal range exists for each steel casting geometry.
In summary, this study underscores the value of numerical simulation in optimizing sand casting processes for steel castings. By combining theoretical calculations with ProCAST simulations, we developed an effective casting plan for the high-pressure steam chamber, minimizing defects and improving yield. The methodology can be extended to other complex steel castings, contributing to advancements in manufacturing efficiency and product quality. Future work will focus on integrating advanced criteria like the Niyama criterion into automated optimization algorithms and exploring additive manufacturing techniques for mold and core production to further enhance steel casting precision.
The successful simulation of this steel casting process demonstrates the critical role of computational tools in modern foundry practices. As steel castings continue to be essential in high-performance applications, such simulations will drive innovation, reduce waste, and ensure reliability. We envision a future where every steel casting design undergoes virtual validation, leading to smarter and more sustainable manufacturing.