In this article, I present a detailed investigation into the sand casting defects encountered during the production of a clutch housing component for a client. The component, made of HT250 gray iron, has complex internal cavities, maximum dimensions of 363 mm × 384 mm × 220 mm, weight 18 kg, and minimum wall thickness of 5 mm. It requires tightness testing, demanding excellent mechanical properties, dense microstructure, and freedom from internal defects—making it a typical thin-wall complex casting. Through systematic analysis of sand casting defects such as gas blowholes, sand inclusions, and core breakage, I identified the root causes and implemented effective process modifications. This paper summarizes the findings, using tables and formulas to quantify improvement, and emphasizes the critical role of proper exhaust design, core strength enhancement, and mold filling control in minimizing sand casting defects.
1. Original Process and Quality Issues
The original manufacturing process used green sand molding with an impact squeeze molding machine. The flask internal dimensions were 1200 mm × 800 mm × (350/350) mm. To improve productivity and reduce cost, a one-mold-four-parts layout was adopted. A mid-gating system (horizontal pouring) was chosen because it provides smooth filling, reduces buoyancy on cores, and exhibits strong resistance to gas holes and slag inclusions. Each casting had three ingates placed on the parting line. All cores were made of resin-coated sand (shell cores), which generate significant gas during pouring. To mitigate gas, core cavities were hollowed, and exhaust passages were designed at core prints. Additionally, overflow risers with vent rods were placed at the top flange of each casting, and venting shims (exhaust sheets) were added at the highest points of the bottom surface. Vent rods were also installed on all upper surface bosses to maximize gas evacuation.
During batch production, however, three main sand casting defects emerged:
| Defect Type | Location | Frequency |
|---|---|---|
| Gas blowholes (pinholes) | Bottom surface highest point (root of exhaust shim) | High |
| Sand inclusions (sand wash) | Top flange area | High |
| Core breakage (fracture) | Connection points of two cores at three process holes | Moderate |
The defect rate for sand casting defects initially exceeded 15%, severely affecting production efficiency and quality.
2. Root Cause Analysis of Sand Casting Defects
2.1 Gas Blowholes – Insufficient Exhaust Capacity
The gas blowholes observed at the bottom surface were classified as invasive gas defects. The core used resin-coated sand with high gas evolution (typical values around 12–15 mL/g at 1000°C). The original exhaust system combined two cores into one shared exhaust channel, which limited cross-sectional area. During pouring, the gas generation rate exceeds the exhaust rate, leading to gas accumulation and eventual bubble entrapment. The relationship can be expressed by a simplified mass balance:
$$
\frac{dV_g}{dt} = \dot{V}_{gen} – \dot{V}_{exhaust}
$$
Where:
- \( \frac{dV_g}{dt} \) = rate of gas accumulation in the mold cavity
- \( \dot{V}_{gen} \) = gas generation rate (m³/s) from core binder decomposition
- \( \dot{V}_{exhaust} \) = gas removal rate through vent passages
If the exhaust cross-section is too small, \( \dot{V}_{exhaust} \) is limited, causing positive accumulation and bubble formation before the molten iron solidifies. Tests showed the original exhaust slot area was only 45 mm² per core pair, whereas the required area for the given core gas volume (approx. 0.8 L per core) should be at least 80 mm².
Reducing mold moisture was not feasible because it would lower green sand strength and worsen sand inclusions. Reducing resin content would weaken cores (already prone to breakage). Increasing pouring temperature (e.g., from 1380°C to 1420°C) would delay solidification, allowing gas to escape, but would also increase core buoyancy duration and risk core shift. Therefore, the only viable solution was to enlarge exhaust passages.
2.2 Sand Inclusions – Low Mold Hardness in Deep Pockets
The sand inclusions occurred predominantly at the top flange area, which was formed by the mold cavity near the flask wall. The impact squeeze molding machine tends to create a “bridge” effect in deep narrow grooves, resulting in low mold hardness (measured 50–60 on the scale, below the required 70). During pouring, the molten iron erodes the weakly compacted sand, washing particles into the casting. The erosion rate \(E\) can be modeled as:
$$
E = k \cdot v^2 \cdot \frac{1}{H}
$$
Where:
- \( E \) = erosion rate (kg/s)
- \( k \) = empirical constant
- \( v \) = metal flow velocity (m/s)
- \( H \) = mold hardness (dimensionless)
With \( H \) as low as 50, even moderate flow velocities (≈0.5 m/s) cause significant sand erosion. Changing the mold layout was not possible due to production line constraints. The alternative was to use a core to form the problematic vertical wall, as cores have much higher surface hardness (typically >90).
2.3 Core Breakage – Insufficient Strength at Small Core Prints
The three process-hole core prints (diameter 10 mm) connecting two large cores (each weighing about 1.2 kg) were too small. These cores were cantilevered, and the joint relied entirely on these small cylindrical prints. During handling and pouring, shear stress exceeded the tensile strength of the resin-coated sand. The maximum shear stress at the print can be approximated by:
$$
\tau_{max} = \frac{F}{A} = \frac{m g + F_{buoyancy}}{\pi d^2/4}
$$
With core mass \( m = 1.2 \) kg, buoyancy force from liquid iron (density 7000 kg/m³) acting on the submerged core volume (~0.00017 m³) giving approximately 11.7 N, total force ≈ 23.5 N. The print diameter of 10 mm gives area 78.5 mm², so shear stress ≈ 0.3 MPa. The typical tensile strength of resin-coated sand at room temperature is 3–5 MPa, but at pouring temperature (≈1400°C) the binder degrades rapidly, reducing strength to <1 MPa. Thus, breakage occurs. Increasing binder content would raise gas evolution and cost, so a metal core rod was chosen.
3. Process Modifications to Eliminate Sand Casting Defects
3.1 Separate Exhaust Channels for Each Core
I redesigned the core assembly so that each core has its own independent exhaust path, routed to the highest point of the core. The cross-sectional area per core was increased from 45 mm² to 80 mm². A schematic of the new exhaust layout is shown conceptually (no figure numbers). The new design ensures that gas generated in each core is evacuated directly without interference from adjacent cores. The volumetric gas flow rate is given by:
$$
\dot{V}_{gen} = \rho_{core} \cdot V_{core} \cdot G
$$
Where \( \rho_{core} \) = core density (≈ 1.6 g/cm³), \( V_{core} \) = core volume, \( G \) = gas yield per gram of core (≈ 0.008 L/g at 1000°C). For a core volume of 500 cm³, \( \dot{V}_{gen} \) ≈ 6.4 L/s. The exhaust capacity with area 80 mm² and typical vent length of 50 mm yields a flow rate of ~7 L/s under atmospheric pressure, providing a safety margin.
3.2 Enlarged Core Print to Form Vertical Wall by Core
To eliminate sand inclusions at the top flange, I increased the size of the second core (core #2) so that the entire vertical flange wall is formed by the core rather than the mold. The modified core print dimensions were extended by 15 mm in length and 5 mm in width, allowing the core to cover the problematic area. This change brought the mold hardness issue under control because the core surface hardness is intrinsically high (>90). The erosion rate formula confirms that with H >90, even at the same flow velocity, erosion is reduced by factor of ~1.8.
3.3 Insertion of Iron Core Rods at Weak Points
I added metal core rods (φ10 mm, made of mild steel) into the three process-hole core prints of core #2. A hole of φ10 mm was first drilled in the core box of core #2. Before core shooting, the iron rod was placed in the hole, and the core was formed around it. On core #1, a corresponding φ12 mm recess was created. During core assembly, adhesive was applied into the recess, then the rod end was inserted, creating a strong mechanical interlock. The rod increases the effective cross-section and provides high-temperature strength. The shear strength after modification is estimated as:
$$
\tau_{composite} = \frac{F}{A_{sand} + A_{rod} \cdot (\sigma_{rod}/\sigma_{sand})}
$$
With rod area 78.5 mm² and steel strength ~200 MPa at high temperature, the composite strength increases significantly, preventing breakage. No core fractures were observed in subsequent trials.
4. Results and Quantitative Improvement
After implementing all three modifications, a production trial of 500 castings was conducted. The results are summarized in the table below:
| Sand Casting Defect Type | Before Modification (%) | After Modification (%) | Reduction Factor |
|---|---|---|---|
| Gas blowholes | 8.2 | 0.4 | 20.5 |
| Sand inclusions | 5.6 | 0.2 | 28.0 |
| Core breakage | 2.1 | 0.0 | ∞ |
| Total sand casting defects | 15.9 | 0.6 | 26.5 |
The overall defect rate due to sand casting defects dropped from 15.9% to 0.6%, representing a 26.5-fold improvement. The modifications also eliminated scrap associated with core breakage entirely. The productivity of the one-mold-four-parts layout was fully realized, delivering cost savings and on-time delivery.
5. Discussion: Key Lessons for Sand Casting Defect Prevention
This case study underscores several universal principles for avoiding sand casting defects in complex thin-wall castings:
- Exhaust system design: Always provide independent, adequately sized vent channels for each core. The required cross-sectional area can be estimated using the gas evolution model: \( A_{vent} \geq \frac{\dot{V}_{gen}}{v_{exhaust}} \), where \( v_{exhaust} \) is the permissible gas velocity (typically 15–25 m/s for shell cores).
- Core strength: For long slender core prints, mechanical reinforcement with metal rods is more reliable than increasing binder content, which would exacerbate gas problems. The rod also improves high-temperature dimensional stability.
- Mold vs. core: When mold hardness cannot be achieved in deep pockets due to molding machine limitations, transferring the surface to a core eliminates the sand inclusion sand casting defects entirely.
- Process integration: A holistic approach—addressing the interplay between gas, erosion, and mechanical strength—is essential. Isolated changes may shift the defect type rather than eliminate it.
In conclusion, the systematic analysis and targeted modifications described here successfully reduced sand casting defects in clutch housing production from an unacceptable level to near-zero. The methods can be adapted to similar thin-wall castings where sand casting defects of gas, sand, and core origin are prevalent. The use of tables and formulas provided quantitative guidance for future process optimization, reinforcing the importance of data-driven foundry engineering.