In our foundry operations, we frequently encounter challenges when producing ductile iron castings with large flat surfaces, particularly when the top surface is expansive. These geometries are prone to various casting defects such as sand inclusions, sand expansion, metal penetration, and gas pores, which can compromise integrity and surface quality. This article shares our firsthand experience and systematic approach to mitigating these casting defects, focusing on practical modifications in molding materials, process design, and pouring techniques. Through iterative trials, we have developed a comprehensive strategy that significantly reduces the incidence of casting defects, ensuring that castings meet stringent technical specifications without substantial cost increases.
The core issue lies in the inherent vulnerabilities of green sand molds when subjected to the high temperatures of ductile iron melt, typically around 1350–1400°C. The thermal shock leads to rapid vaporization of moisture, gas generation, and differential expansion within the mold, fostering an environment conducive to casting defects. Our initial attempts using conventional green sand on a jolt-squeeze molding machine resulted in unacceptable surface defects on the upper plane, despite satisfactory lower surfaces. This prompted a detailed investigation into the root causes and the implementation of tailored countermeasures.
Understanding the Root Causes of Casting Defects
Casting defects in large flat surfaces arise from a complex interplay of factors. We identified several primary mechanisms:
- Gas-Related Defects: The high heat causes moisture in the green sand to vaporize aggressively, generating substantial gases. If venting is inadequate, gases become trapped, leading to侵入性气孔 (invasive pores),析出性气孔 (precipitated pores), or反应性气孔 (reactive pores). The gas pressure $P_g$ can be approximated by the ideal gas law under non-ideal conditions: $$P_g = \frac{nRT}{V} + \Delta P_{surf}$$ where $n$ is moles of gas generated, $R$ is the gas constant, $T$ is temperature, $V$ is volume, and $\Delta P_{surf}$ accounts for surface tension effects in the sand matrix.
- Sand Expansion Defects: Differential thermal expansion between the mold surface and inner layers induces compressive stresses, causing surface layers to buckle or spall—a phenomenon known as sand expansion or scabbing. The thermal stress $\sigma_{th}$ can be modeled as: $$\sigma_{th} = E \cdot \alpha \cdot \Delta T$$ where $E$ is the modulus of elasticity of the sand, $\alpha$ is the coefficient of thermal expansion, and $\Delta T$ is the temperature gradient.
- Sand Inclusion and Metal Penetration: Erosion from molten metal flow can dislodge sand grains, while inadequate mold hardness or high permeability allows metal infiltration into sand interstices.
The following table summarizes common casting defects and their primary causes in our context:
| Type of Casting Defect | Primary Causes | Typical Manifestation |
|---|---|---|
| Gas Porosity (Invasive) | High moisture content, low permeability, poor venting | Round or elongated pores near upper surfaces |
| Sand Expansion (Scabbing) | Low hot strength, high thermal expansion, moisture migration | Flaky layers or lifts on large flat planes |
| Sand Inclusions | Erosion from turbulent flow, weak mold surfaces | Embedded sand particles on casting surface |
| Metal Penetration | High permeability, low mold hardness, high pouring temperature | Rough, adherent metal-sand mixture |

Our analysis revealed that the standard green sand composition, though suitable for many applications, was insufficient for large flat ductile iron castings. The properties are listed below:
| Component/Property | Face Sand | Backing Sand |
|---|---|---|
| New Sand (%) | 30–50 | 0–20 |
| Used Sand (%) | 50–70 | 80–100 |
| Bentonite (%) | 4–5 | 1–1.5 |
| Coal Dust (%) | 3–5 | 0 |
| Soda Ash (%) | 0–0.5 | 0 |
| Heavy Oil (%) | 1–1.5 | 0 |
| Moisture Content (%) | 3.5–4.5 | 3.5–4.5 |
| Green Permeability | ≥ 100 | ≥ 80 |
| Green Compressive Strength (MPa) | 0.06–0.08 | 0.06–0.08 |
| Application | Surface layer | Support layer |
This composition led to a green strength of only 10–20% of dry strength, exacerbating casting defects under thermal load. The mold hardness achieved was around 70%, but gas evolution and expansion issues persisted.
Comprehensive Measures to Prevent Casting Defects
To address these casting defects, we overhauled our process, adopting a surface-dried mold approach combined with倾斜浇注 (tilted pouring). The key was to enhance mold stability, reduce gas generation, and manage thermal gradients. Below, we detail the modifications for each type of casting defect.
1. Mold Material Optimization
We switched to a silica sand with uniform grain size of 0.212–0.425 mm (70–40 mesh), which offers better thermal stability. The binder system was revised to include activated bentonite and wood flour additives. The new composition is summarized in the table:
| Component | Percentage (%) | Function |
|---|---|---|
| Silica Sand (70–40 mesh) | Base | Refractory backbone |
| Bentonite (Activated) | 6–8 | Binder, enhances hot strength |
| Wood Flour | 1–2 | Increases permeability, reduces expansion | Coal Dust | 3–4 | Anti-penetration agent |
| Moisture Content | 4–6 | Controlled for optimal compactibility |
| Green Permeability | ≥ 150 | Measured via standard test |
| Green Compressive Strength (MPa) | 0.10–0.12 | Improved for handling |
| Hot Wet Tensile Strength (kPa) | ≥ 2.5 | Key for resisting sand expansion |
The hot wet tensile strength $S_{hwt}$ is critical for preventing sand expansion defects and can be estimated from bentonite content $C_b$ and moisture $M$: $$S_{hwt} = k_1 \cdot C_b \cdot e^{-k_2 \cdot M} + k_3$$ where $k_1$, $k_2$, $k_3$ are material constants. We aimed for $S_{hwt} > 2.5 \text{ kPa}$ to withstand thermal stresses.
2. Preventing Gas Porosity Casting Defects
Gas-related casting defects were mitigated through multi-faceted actions:
- Enhanced Venting: We added numerous vent holes in the upper mold, especially near large flat areas. The venting capacity $Q_v$ was designed to exceed gas generation rate $G$: $$Q_v = A_v \cdot \sqrt{\frac{2 \Delta P}{\rho}} > G = \frac{dn_g}{dt} \cdot \frac{RT}{P}$$ where $A_v$ is total vent area, $\Delta P$ is pressure differential, $\rho$ is gas density, and $n_g$ is moles of gas.
- Controlled Pouring: Gating systems were redesigned with smooth transitions to ensure laminar flow, reducing turbulence that entraps air. Pouring temperature was lowered to 1320–1340°C, and speed was increased to minimize mold contact time.
- Mold Drying: Surface drying with a kerosene torch reduced surface moisture to below 1%, curtailing immediate gas evolution.
The effectiveness of these measures is evident in the reduction of gas porosity casting defects from ~15% to under 2% in sampled castings.
3. Mitigating Sand Expansion Casting Defects
Sand expansion, a pervasive casting defect in flat planes, was tackled by:
- Increasing Hot Strength: Higher bentonite content and activation with soda ash boosted thermal resistance. We also inserted iron nails into vulnerable upper mold regions to reinforce the sand matrix.
- Reducing Thermal Expansion: Wood flour and coal dust act as buffers, absorbing expansion energy. The effective expansion coefficient $\alpha_{eff}$ is reduced as: $$\alpha_{eff} = \alpha_{sand} – \sum (f_i \cdot \beta_i)$$ where $f_i$ and $\beta_i$ are volume fractions and expansion coefficients of additives.
- Tilted Pouring: By placing the mold on a tilting plate at an angle $\theta = 10–15^\circ$, the upper plane is no longer horizontal, reducing the direct thermal impact area. This alters the heat flux distribution, modeled by: $$q” = k \cdot \frac{\Delta T}{\delta} \cdot \cos(\theta)$$ where $q”$ is heat flux, $k$ is thermal conductivity, $\Delta T$ is temperature difference, and $\delta$ is mold thickness.
4. Eliminating Sand Inclusion and Metal Penetration Casting Defects
To prevent these surface-related casting defects:
- Improved Mold Hardness: We achieved uniform compactibility of 85–90% by optimizing jolt-squeeze cycles and using stiffer sands. Hardness $H$ correlates with reduced metal penetration depth $d_p$: $$d_p \propto \frac{1}{\sqrt{H}}$$
- Anti-penetration Additives: Coal dust forms a gas barrier, while finer sand grains (with a grain fineness number GFN > 70) decrease intergranular spaces.
- Careful Molding: Gates were positioned at the low end of tilted molds to minimize erosion. Cores and molds were cleaned rigorously before closing, and pouring basins were covered to prevent sand fallout.
The combined effect is quantified in the table below, showing defect rates before and after improvements:
| Casting Defect Type | Initial Defect Rate (%) | Post-Improvement Defect Rate (%) | Key Mitigation Factor |
|---|---|---|---|
| Gas Porosity | 12 | 1.5 | Enhanced venting, lower moisture |
| Sand Expansion | 18 | 2.0 | Increased hot strength, tilted pouring |
| Sand Inclusions | 8 | 0.5 | Better mold hardness, gate design |
| Metal Penetration | 10 | 1.0 | Coal dust, higher compactibility |
5. Theoretical Underpinnings and Formulas
Our approach is grounded in foundry science. Key formulas we used to guide decisions include:
- Gas Generation Model: The total gas volume $V_g$ from mold decomposition is: $$V_g = \int_0^t \phi(T(\tau)) \cdot A_m \, d\tau$$ where $\phi$ is gas evolution rate as a function of temperature $T$, $A_m$ is mold area, and $t$ is time.
- Mold Stability Criterion: To avoid sand expansion, the thermal stress must not exceed the hot wet tensile strength: $$\sigma_{th} = E \cdot \alpha \cdot (T_{surface} – T_{inner}) < S_{hwt}$$
- Permeability-Porosity Relation: For sand molds, permeability $k_p$ relates to porosity $\epsilon$ and grain size $d_g$ via the Kozeny-Carman equation: $$k_p = \frac{\epsilon^3}{180 (1-\epsilon)^2} \cdot d_g^2$$ We targeted $k_p > 150$ for adequate venting.
Implementation and Results
We implemented these measures in a production batch of approximately 500 ductile iron plates (size: 250 mm × 150 mm × 25 mm). The process flow involved: sand preparation with controlled parameters, molding on a jolt-squeeze machine to achieve high uniformity, surface drying with torches, placement on tilting plates, and careful pouring with optimized gating. After solidification and cooling, castings were inspected visually and via non-destructive testing.
The results were striking: casting defects were nearly eliminated, with surface roughness $R_a$ improving to below 12.5 μm, meeting specifications. The mechanical properties of ductile iron (tensile strength > 500 MPa, elongation > 10%) remained uncompromised, confirming that our modifications did not induce new casting defects such as shrinkage or misruns. The cost increase was marginal, primarily due to added bentonite and wood flour, offset by reduced scrap rates.
Conclusion
Through a holistic approach addressing mold materials, gating design, and pouring techniques, we have successfully mitigated the persistent casting defects in large flat ductile iron castings. The integration of surface-dried sands, tilted pouring, and enhanced venting proved effective in controlling gas porosity, sand expansion, inclusions, and penetration. Each measure contributed to a robust process where casting defects are minimized without significant trade-offs. This experience underscores the importance of tailored solutions for specific geometries and alloys, and we continue to refine these methods to tackle other challenging casting defects in our foundry operations. The knowledge gained enhances our capability to produce high-integrity castings efficiently, yielding both quality and economic benefits.
