In the production of large-scale pressure vessels, such as valve housings, steel casting is a predominant manufacturing method due to its ability to form complex shapes under high-temperature and high-strength conditions. These steel castings are critical components in industries like power generation and petrochemicals, where reliability and precision are paramount. Valve housings can be classified into thin-walled and thick-walled types based on the ratio K of outer diameter to inner diameter; when K ranges from 1.1 to 1.2, they are considered thin-walled, otherwise thick-walled. The internal cavities of these steel castings are formed using sand cores, with core heads positioned at both ends of the openings. During the pouring process, the buoyant force exerted by molten steel on the sand core is a significant challenge, often reaching 4–5 times the weight of the core. This buoyancy can lead to core floating, a phenomenon where the core shifts upward, resulting in dimensional inaccuracies such as reduced wall thickness in the upper part and increased thickness in the lower part. Such deviations not only cause scrap and rework but also induce defects like shrinkage porosity, compromising the integrity of the steel casting. In this study, I investigate the root causes of core floating in valve housing steel castings and propose comprehensive solutions to enhance core stability, thereby improving the quality and yield of these critical components.
The formation of valve housing steel castings typically involves a horizontal molding approach to facilitate core support and feeding. The mold is divided into two parts: the upper cope and lower drag, with the sand core placed between them to define the internal cavity. However, the core’s stability is jeopardized by the buoyant force during pouring, which can be quantified using the buoyancy formula: $$F_b = \rho_{steel} V_{core} g$$ where $F_b$ is the buoyant force, $\rho_{steel}$ is the density of molten steel (approximately 7,800 kg/m³ for typical steel casting alloys), $V_{core}$ is the volume of the core submerged in the molten steel, and $g$ is the acceleration due to gravity (9.81 m/s²). For a sand core with weight $W_{core} = \rho_{core} V_{core} g$ (where $\rho_{core}$ is the density of the sand core, around 1,600 kg/m³), the ratio of buoyant force to core weight is given by: $$k = \frac{F_b}{W_{core}} = \frac{\rho_{steel}}{\rho_{core}}$$ Substituting values, $k \approx \frac{7800}{1600} = 4.875$, confirming that the buoyant force is 4–5 times the core weight in steel casting processes. This excessive force often overwhelms the core’s anchoring mechanisms, leading to displacement.
To analyze the impact of core floating, I examined typical defects in valve housing steel castings, focusing on dimensional deviations and shrinkage porosity. Dimensional issues arise from uneven wall thickness distribution, while shrinkage porosity results from altered feeding paths due to core movement. For instance, in a large main steam valve housing steel casting with轮廓 dimensions of 2,990 mm × 2,200 mm × 1,550 mm and a weight of approximately 18 tons, the thin-walled structure (minimum wall thickness 100 mm) exhibited significant漂芯. Measurements from similar steel castings (Valve Housings 1–6) revealed consistent patterns of wall thickness variation, as summarized in Table 1. The data indicate that as the core span increases, the漂芯量 rises, highlighting the correlation between core stiffness and stability in steel casting production.
| Steel Casting | Core Span L | Core Rod Diameter | Theoretical Wall Thickness | Actual Wall Thickness Measurements | Floating Core Amount | ||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| a | b | c | d | e | f | g | |||||
| Valve Housing 1 | 2,700 | 100 | 90 | 75 | 81 | 90 | 98 | 105 | 97 | 82 | 15 |
| Valve Housing 2 | 2,700 | 100 | 90 | 73 | 80 | 89 | 99 | 107 | 98 | 81 | 17 |
| Valve Housing 3 | 2,700 | 100 | 90 | 74 | 81 | 91 | 97 | 106 | 97 | 82 | 16 |
| Valve Housing 4 | 3,590 | 120 | 100 | 77 | 87 | 99 | 110 | 122 | 110 | 86 | 23 |
| Valve Housing 5 | 3,590 | 120 | 100 | 80 | 89 | 100 | 110 | 120 | 109 | 88 | 20 |
| Valve Housing 6 | 3,590 | 120 | 100 | 79 | 88 | 100 | 111 | 121 | 110 | 87 | 21 |
The dimensional deviations in these steel castings are accompanied by shrinkage porosity, detected through non-destructive testing (NDT). Although simulation software predicted no shrinkage, the actual defects emerged due to core floating, which modifies the solidification dynamics. The relationship between core displacement and shrinkage can be modeled using the feeding efficiency equation: $$V_{shrinkage} = V_{casting} – V_{fed}$$ where $V_{shrinkage}$ is the volume of shrinkage porosity, $V_{casting}$ is the total volume of the steel casting, and $V_{fed}$ is the volume fed by risers. When the core floats, the wall thickness changes alter the thermal gradients, reducing $V_{fed}$ and increasing $V_{shrinkage}$. This underscores the necessity of addressing core stability in steel casting to prevent both dimensional and internal defects.
To mitigate core floating, I implemented several改进措施 targeting core rigidity, fixation, and process control. The first approach involved redesigning the core rod from a single-row structure to a double-row configuration. The original core rod, made of ϕ100–120 mm round steel in a single layer, lacked sufficient stiffness for large-span cores in steel casting. The new double-row core rod, as illustrated in Figure 5, consists of upper and lower main trunks (1) welded into a T-shape using直径 100–120 mm runner bars (repurposed to reduce costs), cross ties (2) of ϕ20–30 mm round steel spaced 500–600 mm apart, and stabilizing ties (3) of 60–80 mm square steel. The bending stiffness $EI$ of a beam is proportional to the moment of inertia $I$, and for a double-row arrangement, the effective $I$ increases significantly. For a rectangular cross-section, $I = \frac{bh^3}{12}$, where $b$ is width and $h$ is height. By doubling the rows, the height $h$ effectively doubles, leading to an eightfold increase in stiffness: $$I_{double} \approx 8 I_{single}$$ This enhancement resists deformation under buoyant forces in steel casting.
Furthermore, I introduced a core head fixation method to secure the sand core to the lower mold. During molding, pre-embedded standard cores (2) of 60–80 mm square steel are placed in the lower drag (1), while corresponding standard cores (4) are welded to the core rod (3) at the core head locations. After core setting, these elements are welded together, ensuring three-point fixation that anchors the core against buoyancy. The fixation force $F_{fix}$ must satisfy: $$F_{fix} \geq F_b – W_{core}$$ where $F_b$ is the buoyant force and $W_{core}$ is the core weight. For typical steel casting conditions, $F_{fix}$ is designed to exceed the net upward force, preventing位移.
Another critical改进 is the core head间隙 inspection using clay strips. After core setting, dual clay strips (2) of ϕ8–10 mm diameter are applied on each core head (1). The upper mold is then closed and reopened to assess the compression of the clay strips. The间隙量 $\delta$ is calculated as: $$\delta = t_{initial} – t_{compressed}$$ where $t_{initial}$ is the initial clay strip thickness (8–10 mm) and $t_{compressed}$ is the thickness after compression. The standard工艺 requires $\delta$ to be 1–3 mm. If $\delta > 3$ mm, shims of thin steel plates are added to eliminate excess间隙, ensuring full contact and stability during steel casting pouring.
Additionally, I applied a反变形量 to compensate for core floating. By adding material to the upper part of the core box based on the upper tolerance limit, the internal cavity is pre-distorted to offset anticipated displacement. The反变形量 $\Delta$ can be estimated from historical漂芯 data: $$\Delta = \alpha L$$ where $\alpha$ is a coefficient derived from regression analysis of previous steel casting measurements (e.g., $\alpha \approx 0.006$ for spans around 3,000 mm), and $L$ is the core span. This proactive approach reduces post-casting adjustments in steel casting production.
To validate these改进措施, I conducted trials on multiple valve housing steel castings and compared the results with prior data. The effectiveness was evaluated using dimensional accuracy and defect reduction metrics, as shown in Table 2. The floating core amount decreased dramatically, and shrinkage porosity volume was minimized, leading to lower rework rates. The焊补率, defined as the volume of weld repair per ton of steel casting, improved significantly, demonstrating the overall enhancement in quality for steel casting components.
| Parameter | Before Improvement | After Improvement |
|---|---|---|
| Floating Core Amount (mm) | 10–25 | 1–3 |
| Shrinkage Porosity Volume (dm³) | 2.5–3.0 | 0.3–0.5 |
| Weld Repair Rate (dm³/t) | 1.64 | 0.24 |
The reduction in floating core amount can be modeled using the stiffness improvement factor. For a double-row core rod, the deflection $\delta$ under a uniformly distributed buoyant load $q$ (force per unit length) is given by: $$\delta = \frac{q L^4}{384 E I_{double}}$$ Compared to the single-row core rod with stiffness $I_{single}$, the deflection ratio is: $$\frac{\delta_{double}}{\delta_{single}} = \frac{I_{single}}{I_{double}} \approx \frac{1}{8}$$ Thus, the core displacement is reduced by about 87.5%, aligning with the observed 1–3 mm漂芯量 in improved steel castings. This theoretical analysis confirms the efficacy of the double-row design in steel casting applications.
Moreover, the decrease in shrinkage porosity volume $V_{shrinkage}$ can be attributed to more consistent wall thickness, which optimizes feeding. Using Chvorinov’s rule, the solidification time $t_s$ is: $$t_s = C \left( \frac{V}{A} \right)^2$$ where $C$ is a mold constant, $V$ is volume, and $A$ is surface area. Core floating causes variations in $V/A$, leading to isolated hot spots. By stabilizing the core, $V/A$ becomes uniform, reducing $t_s$ variations and minimizing shrinkage in steel casting. The improved weld repair rate reflects fewer defects, lowering production costs and cycle times for steel casting foundries.
In conclusion, this study addresses the persistent issue of core floating in valve housing steel castings through a multifaceted approach. The double-row core rod design enhances stiffness economically by utilizing runner bars, while core head fixation and间隙 inspection ensure stability during molding. The反变形量 compensates for anticipated displacements, and together, these measures reduce floating core amounts to within tolerance limits (1–3 mm) and shrinkage porosity volumes to 0.3–0.5 dm³. The weld repair rate drops from 1.64 dm³/t to 0.24 dm³/t, indicating substantial quality improvement in steel casting production. Future work could explore advanced materials for core rods or real-time monitoring systems to further optimize steel casting processes. By implementing these solutions, manufacturers can achieve higher yields, reduced scrap, and enhanced reliability for critical steel casting components like valve housings, contributing to the overall advancement of the steel casting industry.
The success of this research underscores the importance of integrated solutions in steel casting technology. From design to process control, every step must be meticulously planned to counteract forces like buoyancy. The formulas and tables presented here provide a framework for analyzing and addressing core floating, applicable not only to valve housings but also to other large, hollow steel castings. As demand for high-performance steel castings grows, continued innovation in core stability will be essential for meeting stringent quality standards and driving efficiency in steel casting operations worldwide.