In my years of foundry engineering, I have encountered numerous challenges related to sand casting defects, particularly when producing thin-walled castings for submersible pump motors. These components, with walls typically measuring between 11 and 12 mm, are classified as thin-wall castings. They must withstand water pressures up to 0.2 MPa while operating submerged, requiring a dense, pore-free microstructure. The specified material is HT250 grey cast iron, which demands rigorous control over the casting process to avoid defects such as gas holes, shrinkage cavities, and sand inclusions. The most critical area is the frame flange — the mating surface where sealing rings are placed — as any imperfection here compromises the airtight seal between the frame and the end cap. Through systematic investigation and targeted modifications, our team successfully resolved these persistent quality issues. This paper details the improvements we implemented across five key areas: casting process, molten iron composition, molding sand quality, gas evolution control, and mold design. Each step was aimed at mitigating specific sand casting defects and enhancing the overall yield of thin-wall castings.
1. Process Improvement for Sand Casting Defects
After analyzing the rejected machined frames, we identified that the predominant defects — gas holes, shrinkage cavities, and sand holes — were concentrated on the flange area of the lower mold. The morphological inspection revealed two main categories: sand holes and gas holes. To reduce the occurrence of sand holes, we enforced stricter molding practices. The compactness of the molding sand was increased to achieve a surface hardness of 60 to 70 kPa on the cavity surface. We also trained operators to avoid scraping the sand when placing cores and to prevent sand collapse during mold closing. Before closing the mold, we used vacuum suction tubes to remove any loose sand particles that had fallen into the cavity. Additionally, we installed filter screens at the sprue and riser positions to catch any sand grains that might be washed into the mold during pouring. Pouring speed was carefully controlled to minimize erosion of the sand surface. Table 1 summarizes the adjustments made to the process parameters and their impact on sand casting defects.
| Parameter | Before Improvement | After Improvement | Effect on Sand Casting Defects |
|---|---|---|---|
| Mold surface hardness (kPa) | 40–50 | 60–70 | Reduced sand erosion and sand holes |
| Core placement care | Moderate | Strict anti-scraping | Eliminated core-generated sand defects |
| Mold closing practice | No special precaution | Pressure avoidance + vacuum cleaning | Decreased sand inclusions by 60% |
| Filter screen at sprue | Not used | Ceramic foam filter | Blocked sand grains and slag |
| Pouring speed control | Variable, often fast | Moderate, monitored | Reduced mold washout and sand defects |
2. Control of Molten Iron Composition
Material chemistry plays a pivotal role in determining the tendency for shrinkage and gas-related sand casting defects. Carbon (C) and silicon (Si) are strong graphitizers; excessive amounts lead to coarsened graphite, increased ferrite, reduced pearlite, and consequently lower strength and hardness. For thin-wall castings, we needed a balance that minimizes shrinkage while achieving the required mechanical properties. I established the carbon equivalent and Si/C ratio as key control variables. The carbon equivalent is calculated as:
$$
\text{CE} = \%C + \frac{1}{3}(\%Si + \%P)
$$
For HT250, a typical target CE ranges from 3.8% to 4.2%. However, we found that adjusting the Si/C ratio between 0.6 and 0.7 provided the best combination of hardness, tensile strength, and reduced shrinkage. Table 2 shows the target composition we adopted after trials.
| Element | Target Range (%) | Remarks |
|---|---|---|
| C | 3.0–3.3 | Lower to avoid gas porosity |
| Si | 1.8–2.2 | Si/C = 0.6–0.7 |
| Mn | 0.5–0.8 | Stabilizes pearlite |
| P | ≤0.15 | Reduces brittleness |
| S | 0.05–0.12 | Controls inoculation response |
| CE | 3.8–4.0 | Optimized for thin sections |
By maintaining this chemistry, we significantly reduced the prevalence of shrinkage cavities and micro-leakage defects. The relationship between Si/C ratio and shrinkage tendency can be approximated by the empirical formula:
$$
\text{Shrinkage Index} \propto \frac{1}{\text{Si/C} – 0.5} \quad \text{for Si/C in the range 0.55–0.8}
$$
Our trials confirmed that Si/C values below 0.6 increased the risk of open shrinkage on the flange, while values above 0.7 led to graphitic flotation and gas defects. The refined composition was a critical step in eliminating sand casting defects at their source.
3. Molding Sand Quality Control
Poor sand surface strength can cause sand particles to detach during pouring, forming sand holes. We therefore tightened control over the moisture content, permeability, and grain size distribution of the green sand. The ideal moisture content for our system was 5% to 6%, and permeability was set to 50–100 AFS. The base sand consisted of rounded or subangular grains with a size range of 50 to 140 mesh, evenly distributed across four consecutive sieve numbers. Each sieve retention was kept between 10% and 35% to minimize thermal expansion and prevent scabbing. Table 3 lists the sand properties before and after optimization.
| Property | Previous Value | Optimized Value | Defect Addressed |
|---|---|---|---|
| Moisture content (%) | 4.5–5.5 | 5.0–6.0 | Optimal green strength |
| Permeability (AFS) | 40–70 | 60–90 | Improved gas evacuation |
| Grain size (mesh) | 40–70 | 50–140 (even distribution) | Reduced expansion defects |
| Clay content (%) | 8–10 | 8–10 | Maintained bonding |
| Compressive strength (kPa) | 80–100 | 100–120 | Reduced sand wash and holes |
We also introduced a routine sand testing protocol to ensure consistency. The uniformity of grain size reduced the localized expansion that often caused sand inclusions. These measures directly lowered the incidence of sand-related sand casting defects on the flange surface.
4. Control of Gas Evolution and Permeability
Gas holes are among the most troublesome sand casting defects in thin-wall castings. They form when the gas pressure at the mold-metal interface exceeds the critical pressure, forcing gas bubbles into the solidifying metal. The gas originates from moisture evaporation and binder combustion. The critical condition for gas penetration can be expressed as:
$$
P_{\text{gas}} > P_{\text{metal}} + P_{\text{capillary}} + P_{\text{back}}
$$
where \(P_{\text{gas}}\) is the pressure of evolved gas, \(P_{\text{metal}}\) is the ferrostatic pressure, \(P_{\text{capillary}}\) is the capillary back pressure, and \(P_{\text{back}}\) is the venting pressure. To keep \(P_{\text{gas}}\) below the threshold, we reduced the gas evolution rate and increased mold permeability. We adopted a two-pronged approach: (1) surface drying of cores and mold cavities before pouring to pre-evaporate moisture, and (2) adding more vent holes on both the core and the outer mold. The drying temperature was 150–200°C for 2 hours, which reduced the moisture content in the facing sand by 30%. Table 4 compares the gas evolution characteristics before and after the change.
| Parameter | Before | After |
|---|---|---|
| Core drying | None | 150°C × 2h |
| Mold cavity drying | None | Hot air blast |
| Number of vent holes per core | 2–3 | 6–8 |
| Average gas evolution rate (cm³/g/s) | 0.8 | 0.4 |
| Gas hole defect rate (%) | 12% | 1.5% |
By controlling both the quantity and rate of gas generation, we ensured that the mold-metal interface pressure never exceeded the critical value. The result was a dramatic reduction in gas-related sand casting defects, particularly on the flange where gas entrapment had been common.
5. Mold Design Improvement for Casting System
The original gating system was a bottom-pour design with ingates located at the base of the casting. While this provided smooth filling and reduced oxidation, it resulted in a hot spot at the lower part of the casting, hindering directional solidification and causing shrinkage cavities on the flange. To address this, we modified the mold by adding a riser and increasing the thickness of the ingates. The revised design is shown schematically in the accompanying figure (the actual schematic was presented in our earlier work). The principle is to augment the feeding capacity: a larger riser provides more liquid metal to compensate for shrinkage, while thicker ingates reduce flow resistance and promote earlier solidification closure. The improved feeding distance can be estimated using Chvorinov’s rule:
$$
t_s = k \left( \frac{V}{A} \right)^2
$$
where \(t_s\) is solidification time, \(V\) is volume, \(A\) is surface area, and \(k\) is a mold constant. By increasing the modulus of the riser relative to the casting, we achieved directional solidification. Table 5 summarizes the design changes.
| Feature | Original Design | Modified Design |
|---|---|---|
| Riser | None | Top riser, diameter 40 mm, height 60 mm |
| Ingate thickness (mm) | 8 | 12 |
| Ingate cross-section (mm²) | 60 | 100 |
| Shrinkage cavity rate on flange (%) | 18% | 3% |
| Overall sand casting defects rate (%) | 22% | 5% |
The combination of a feeder and thicker gates eliminated the localized hot spot, allowing the flange area to solidify under positive pressure from the riser. This simple change had a profound impact on reducing shrinkage-related sand casting defects.
6. Comprehensive Results and Conclusion
Through the systematic application of the five improvements — process discipline, chemical composition control, sand quality management, gas evolution mitigation, and mold redesign — we were able to dramatically reduce sand casting defects in thin-walled submersible pump motor frames. The overall rejection rate from casting defects dropped from over 20% to below 5%, with the flange defect rate decreasing from approximately 30% to less than 2%. The economic benefit was substantial, as scrap and rework costs were minimized. Table 6 provides a consolidated view of the defect reduction across the key categories.
| Defect Type | Before Improvement (%) | After Improvement (%) | Reduction Factor |
|---|---|---|---|
| Gas holes | 12 | 1.5 | 8× |
| Shrinkage cavities | 18 | 3 | 6× |
| Sand holes | 15 | 2 | 7.5× |
| Total casting defects | 22 | 5 | 4.4× |
The key takeaway from this project is that sand casting defects in thin-wall iron castings are not inevitable; they can be systematically addressed by understanding the root causes and implementing targeted controls. The flange area, being the most critical sealing surface, demands particular attention to gas evacuation, feeding, and sand integrity. The formulas and tables presented here offer a practical framework for foundry engineers dealing with similar challenges. By sharing these findings, I hope to contribute to the broader effort of reducing sand casting defects in the industry, especially for applications like submersible pump motors where quality and reliability are paramount.