In our work at a railway casting facility, we faced a persistent challenge with sand casting defects, specifically scabbing and sand inclusion, in the production of ductile iron bearing bushings (QT500-7, each weighing about 9 kg). The vertical flaskless shoot-squeeze molding line produced a qualification rate of only about 60%, with scabbing defects accounting for more than 70% of all rejects. These sand casting defects appeared on the arc surface and vertical planes of the bushing, ranging from 3 to 7 mm in depth and varying in area. Overcoming these sand casting defects was the key to improving product yield. In this article, I will share the process measures we implemented to eliminate sand casting defects in bearing bushings, focusing on two main aspects: modifying the gating system and enhancing the green sand properties through bentonite activation.
We began by analyzing the root causes of these sand casting defects. The original gating system was a closed type, with the ingate directed straight at the sand mold cavity wall, as shown in the schematic below. This caused severe thermal shock on the local sand surface, leading to overheating and subsequent expansion. On the other hand, the molding sand quality was insufficient; its hot-wet tensile strength was too low to withstand the thermal stress during pouring. As a result, the sand surface cracked, bulged, and peeled off, forming sand casting defects. This analysis guided our two-pronged approach: first, to reduce the direct thermal impact of molten iron on the sand, and second, to improve the high-temperature strength of the sand by modifying the bentonite binder.
Modification of the Gating System
To mitigate the direct thermal shock of molten metal on the sand mold, we experimented with three different pouring methods based on the principle of reducing the heat flux intensity at the mold surface. The first approach involved redirecting the ingate so that the metal stream entered tangentially along the curved surface of the cavity. However, due to the confined geometry of the bearing bushing cavity, this modification yielded only marginal improvement in sand casting defects. The second attempt used a bottom-pour with top-riser system, but the excessively offset sprue caused a significant temperature drop of the molten iron, which actually increased the defect rate. Finally, we adopted a top-pour system as illustrated in the figure. This design ensured both adequate filling speed and avoidance of direct metal impingement on the cavity walls. The result was a reduction in the total reject rate from 40% to about 30%, but sand casting defects still persisted. It became clear that gating improvements alone could not fully eliminate the problem; we needed to strengthen the sand itself.
Enhancing Mold Sand Properties via Bentonite Activation
During pouring of bearing bushings, the local surface of the sand mold is rapidly heated, causing moisture to migrate inward and accumulate in a layer 3–7 mm below the surface. This concentrated moisture zone drastically weakens the sand’s strength. When the internal stress from thermal expansion exceeds the reduced strength, cracks and bulges form, leading to sand casting defects. The original sand mixture used a natural calcium bentonite from Weifang as the binder. Calcium bentonite has a lower swelling capacity and inferior thermal stability compared to sodium bentonite. The key mineral is montmorillonite, which has a structure consisting of two silica tetrahedral sheets sandwiching an alumina octahedral sheet. The unit layers are held together by weak van der Waals forces, allowing water molecules to enter and cause interlayer expansion. Moreover, isomorphous substitution in the lattice gives the montmorillonite surface a net negative charge, attracting exchangeable cations such as Ca²⁺ and Na⁺.
The hydration energy of Na⁺ is significantly higher than that of Ca²⁺, so Na⁺ can replace Ca²⁺ in the interlayer, transforming calcium bentonite into sodium bentonite. This process, called activation, greatly increases the swelling volume and improves the hot-wet tensile strength of the molding sand. We used sodium carbonate (Na₂CO₃) as the activating agent. The chemical reaction can be represented as:
$$ \text{Ca-montmorillonite} + \text{Na}_2\text{CO}_3 \rightarrow \text{Na-montmorillonite} + \text{CaCO}_3 \downarrow $$
The degree of activation depends on the amount of Na₂CO₃ added. In our production trials, we found that adding 3–5% Na₂CO₃ (by weight of bentonite) yielded optimal results. The activated bentonite showed marked improvements in key performance indicators, as summarized in Table 1.
| Property | Unit | Non-activated Ca-bentonite | Activated Na-bentonite (3% Na₂CO₃) |
|---|---|---|---|
| Swelling volume (24h, distilled water) | mL/g | 15 | 35 |
| Methylene blue adsorption (MBV) | g/100g | 32.8 | 34.2 |
| Green compressive strength (of sand mix) | kPa | 85 | 105 |
| Dry compressive strength (of sand mix) | kPa | 440 | 510 |
| Hot-wet tensile strength (at 600°C) | kPa | 1.2 | 3.8 |
Table 1 clearly shows that activated bentonite not only increases swelling volume but also enhances both green and dry strength of the sand mixture. Most importantly, the hot-wet tensile strength—a critical parameter for resisting sand casting defects—nearly tripled. This dramatic improvement can be attributed to the higher water retention and binding capacity of sodium bentonite at elevated temperatures. We also measured the methylene blue adsorption (MBV) of the bentonite after calcination at various temperatures to evaluate thermal stability (Table 2).
| Temperature (°C) | Non-activated Ca-bentonite MBV (g/100g) | Activated Na-bentonite MBV (g/100g) |
|---|---|---|
| Room temp | 32.8 | 34.2 |
| 300 | 23.0 | 27.5 |
| 400 | 2.2 | 7.4 |
| 600 | <0.1 | 3.2 |
| 700 | <0.1 | <0.1 |
Table 2 demonstrates that after calcination at 600°C, the non-activated calcium bentonite lost almost all its adsorption capacity, indicating complete structural collapse. In contrast, the activated sodium bentonite still retained a measurable MBV of 3.2 g/100g, signifying partial preservation of the montmorillonite lattice. This enhanced thermal stability directly reduces the risk of sand casting defects because the sand surface maintains sufficient strength during the critical early stages of solidification.
Theoretical Basis for Hot-Wet Tensile Strength Improvement
The mechanism by which activated bentonite reduces sand casting defects can be explained by the relationship between the swelling index and the cohesive forces in the green sand. The swelling index \( S \) of a bentonite is defined as the volume increase when 1 g of dry bentonite is dispersed in water for 24 h. For calcium bentonite, \( S_{\text{Ca}} \approx 15 \; \text{mL/g} \). For sodium bentonite, \( S_{\text{Na}} \approx 35 \; \text{mL/g} \). The higher swelling capacity means that the activated bentonite forms a more extensive and viscous gel around the sand grains. This gel layer provides greater resistance to moisture migration during rapid heating. The critical hot-wet tensile strength \( \sigma_{\text{HWT}} \) can be modeled by the following empirical equation:
$$ \sigma_{\text{HWT}} = k \cdot \left( \frac{S}{S_0} \right)^n \cdot \exp\left(-\frac{E_a}{RT}\right) $$
where \( k \) is a constant dependent on sand grain distribution, \( S_0 \) is a reference swelling index (e.g., 20 mL/g), \( n \) is an exponent typically between 0.5 and 1.0, \( E_a \) is the activation energy for moisture evaporation, and \( RT \) has its usual meaning. With activated bentonite, \( S \) increases by a factor of 2.3, so \( \sigma_{\text{HWT}} \) roughly doubles (assuming \( n \approx 1 \)). This matches our experimental observations where the hot-wet strength rose from 1.2 kPa to 3.8 kPa.
Furthermore, the activation process also influences the rate of thermal decomposition of the bentonite. The kinetic equation for the loss of structural hydroxyl groups can be expressed as:
$$ \frac{d\alpha}{dt} = A \exp\left(-\frac{E}{RT}\right) (1-\alpha)^m $$
where \( \alpha \) is the fraction decomposed, \( A \) is the pre-exponential factor, \( E \) is the activation energy, and \( m \) is the reaction order. Our thermogravimetric analysis (not shown here) indicated that sodium bentonite has a higher activation energy for dehydroxylation, meaning it requires more thermal energy to break down. As a result, the sand mold retains its binding strength longer when exposed to the intense heat from the molten iron, thereby suppressing scabbing and sand casting defects.
Production Implementation and Results
We implemented the modified gating system (top-pour) along with the use of activated bentonite in the sand mix. The bentonite addition was reduced from 8% (original) to 3–5% (after activation) of the sand weight, because the activated material had higher efficiency. The production process flow is summarized as follows:
| Step | Description | Parameter |
|---|---|---|
| 1. Bentonite activation | Mix calcium bentonite with 3–5% Na₂CO₃ powder and water (30% moisture), age for 24 h. | Temperature: 20–25°C; water addition: 30% of bentonite weight |
| 2. Sand preparation | Add activated bentonite (3–5% of sand) + coal dust (5%) + water (3.5%) to silica sand (AFS 55). | Mull time: 3 min; moisture content: 3.5% |
| 3. Molding | Vertical flaskless shoot-squeeze machine; mold hardness 85–90. | Shoot pressure: 0.6 MPa; squeeze pressure: 1.0 MPa |
| 4. Pouring | Top-pour gating system; pouring temperature 1380–1420°C; pouring time 8–10 s. | Casting yield: 65% |
| 5. Shakeout | After 30 min cooling, shakeout and shot blast. | — |
The results were dramatic. The reject rate due to sand casting defects dropped from over 40% to below 7%. The overall qualification rate of bearing bushings increased from 60% to 93%. Table 3 summarizes the comparison before and after the improvements.
| Performance Indicator | Before (original process) | After (improved process) |
|---|---|---|
| Reject rate from sand casting defects | 40% (approx. 70% of total rejects) | <7% (only minor defects) |
| Total qualification rate | 60% | 93% |
| Bentonite usage | 8% of sand weight | 3–5% of sand weight |
| Hot-wet tensile strength of sand | 1.2 kPa | 3.8 kPa |
| Swelling volume of bentonite | 15 mL/g | 35 mL/g |
We also observed that the sand casting defects completely disappeared from the critical arc surface of the bearing bushing. The few remaining defects were small surface pits caused by slight moisture condensation, easily removed by grinding. The cost of adding Na₂CO₃ was negligible compared to the savings from reduced scrap and improved productivity.
Discussion and Conclusion
Through systematic experimentation, we confirmed that a combined approach of modifying the gating system and activating the bentonite binder is the most effective way to eliminate sand casting defects in bearing bushings. The gating modification alone reduced the thermal shock but was insufficient; the fundamental solution lay in improving the sand’s high-temperature behavior. Activation of calcium bentonite with 3–5% Na₂CO₃ produced a sodium-exchanged bentonite with superior swelling capacity, thermal stability, and hot-wet tensile strength. This allowed the sand skin to resist the thermal expansion forces during pouring, thereby preventing the formation of scabs and sand inclusion.
Key equations that govern the process include the swelling index relationship and the kinetic stability of the binder:
The activation reaction:
$$ \text{Ca-montmorillonite} + \text{Na}_2\text{CO}_3 \rightarrow \text{Na-montmorillonite} + \text{CaCO}_3 $$
The hot-wet strength model:
$$ \sigma_{\text{HWT}} \propto S^n \, \exp\left(-\frac{E_a}{RT}\right) $$
And the decomposition kinetics:
$$ \frac{d\alpha}{dt} = A\,e^{-E/RT}(1-\alpha)^m $$
These fundamental principles guided our process optimization. Our production data confirm that the measures completely resolved the chronic sand casting defects, raising the qualification rate from 60% to over 93%. Although the modification introduced an additional step of bentonite activation, the overall cost was lower because less binder was required and scrap was drastically reduced. We recommend this method for any foundry facing similar sand casting defects in ductile iron castings produced on vertical flaskless molding lines.
In summary, the combination of top-pour gating (to reduce direct thermal impact) and bentonite activation (to increase hot-wet strength) systematically eliminated sand casting defects in bearing bushings. The simple, cost-effective approach can be easily implemented in industrial settings and provides both quality and economic benefits.