In my work on high-pressure steam turbine components, I encountered a particularly challenging box casting made from 06Cr13Ni4Mo stainless steel. This component is critical for the turbine’s high-pressure section, and it must withstand severe temperature and pressure conditions. The design requires extremely high surface and internal quality, with stringent non-destructive testing requirements for steam inlet areas and flange surfaces. The casting has a wall thickness of 32 mm, a central large cavity, and four smaller cavities, making the feeding and core positioning extremely difficult. The geometry is complex, and if the process is not carefully designed, issues with fluidity, wall thickness uniformity, and cleaning will arise, ultimately compromising casting quality. During initial production, I observed several sand casting defects that led to scrapping: cold shuts on local surfaces, shrinkage porosity on surfaces, and large areas of subsurface blowholes. These defects are classic manifestations of sand casting defects in stainless steel castings, and they required systematic analysis and process improvement.
Root Cause Analysis of Sand Casting Defects
Stainless steel 06Cr13Ni4Mo has high chromium content, which significantly reduces its fluidity compared to low-alloy steels. This poor fluidity predisposes the casting to cold shuts and surface wrinkling. In the original process design, I used waist-shaped open risers exclusively at thick sections, but the gating system was not distributed evenly. As a result, during filling, local metal flow was sluggish, leading to inadequate mold filling and poor feeding, which manifested as cold shuts on the casting surface. This is a common sand casting defect in complex stainless steel components.
Subsurface blowholes appeared with higher frequency than in typical low-alloy steel castings. The sand mold’s venting was insufficient, and gases generated during pouring could not escape effectively. These gases penetrated the casting surface and formed blowholes beneath the skin. This type of sand casting defect is particularly troublesome because it is often only discovered after machining, leading to rework or scrap.
Shrinkage porosity and macro-shrinkage occurred at the last solidifying regions. The volumetric shrinkage of stainless steel is higher than that of low-alloy steels, and the lateral feeding capability of the risers was weak. The effective feeding distance of the risers was too short, so hot spots in the casting could not be properly compensated. This resulted in internal sand casting defects that degraded mechanical properties and caused failure in pressure testing.
Numerical Simulation and Process Optimization
To overcome these sand casting defects, I employed numerical simulation software to analyze the filling and solidification behavior. This approach allowed me to predict defect locations and thermal gradients before making physical modifications. I optimized the gating system and riser parameters: I changed the original layout to include both waist-shaped open risers and waist-shaped blind top risers. For the thick sections, I added chills to increase local cooling rates and improve feeding. I increased the size of the No.1 riser, converted the No.4 waist-shaped open riser into a waist-shaped blind top riser, and changed the No.3 cylindrical open riser into a waist-shaped open riser while reducing its height. To enhance venting and eliminate blowholes, I placed ingates at the No.4 blind riser and added venting cords and exhaust holes at critical locations. The entire cavity was defined by a single core (Core No.1) with accurate core prints for positioning.
I used the simulation to map temperature fields and liquid fraction distributions during solidification. This enabled me to adjust the thermal gradients between the hot spots and the risers, ensuring directional solidification. The modified process parameters are summarized in the following table:
| Parameter | Original Design | Optimized Design |
|---|---|---|
| Riser type at thick sections | Waist-shaped open risers | Mix of waist-shaped open and blind top risers |
| No.1 riser size | Standard | Increased by 20% |
| No.4 riser type | Waist-shaped open | Waist-shaped blind top |
| No.3 riser type | Cylindrical open | Waist-shaped open (height reduced 15%) |
| Chills | Not used | Placed at thick walls |
| Gating location | Distributed evenly | Added ingates at No.4 blind riser |
| Venting | Insufficient | Venting cords and exhaust holes added |
The feeding distance of a riser can be estimated using the modulus method. For a plate-like section, the feeding distance \( L \) is given by:
$$ L = k \cdot M $$
where \( M \) is the modulus (volume/surface area) and \( k \) is a constant depending on the alloy. For stainless steel, \( k \) is approximately 2.5. To ensure complete feeding, the feeding distance must be less than twice the riser diameter for open risers. In the optimized design, I reduced the hot spot modulus by adding chills, which increased the effective feeding distance. The solidification shrinkage percentage \( \varepsilon \) for 06Cr13Ni4Mo is about 4.5%, which is higher than typical low-alloy steels. The required riser volume \( V_r \) can be estimated from the casting volume \( V_c \) and shrinkage:
$$ V_r = \frac{\varepsilon \cdot V_c}{1 – \varepsilon + \eta} $$
where \( \eta \) is the riser efficiency (typically 0.14 for open risers and 0.20 for blind risers). The optimized design used blind risers with higher efficiency, reducing the required riser volume while improving feeding.
Implementation Results and Quality Improvement
After implementing the optimized process, I produced one box casting. The results were dramatic. The cold shuts disappeared completely because the improved gating system provided uniform and rapid filling of the mold cavity. The subsurface blowholes were eliminated due to the enhanced venting system, which allowed gases to escape before the metal solidified. Shrinkage porosity and macroshrinkage were absent in the critical thick sections, as confirmed by ultrasonic and radiographic inspection. The casting passed all dimensional and mechanical tests, achieving a 100% first-pass yield compared to the previous 0% yield.
The following table summarizes the defect occurrence before and after optimization:
| Defect Type | Before Optimization (1 casting) | After Optimization (1 casting) |
|---|---|---|
| Cold shuts | Present (scrapped) | None |
| Surface shrinkage porosity | Present | None |
| Subsurface blowholes | Large areas present | None |
| Internal shrinkage (macro/micro) | Present at hot spots | None |
| Overall casting quality | Rejected | Accepted, 100% compliant |
Through numerical simulation, I was able to predict and eliminate the root causes of these sand casting defects. The simulation allowed me to visualize the temperature gradients during solidification. The thermal field in the optimized design promoted directional solidification toward the risers. The temperature gradient \( G \) along the casting length can be expressed as:
$$ G = \frac{T_{\text{riser}} – T_{\text{cold end}}}{L} $$
where \( T_{\text{riser}} \) is the temperature at the riser-casting interface, \( T_{\text{cold end}} \) is the temperature at the far end, and \( L \) is the distance. A sufficiently high gradient ensures that liquid metal flows from the riser to feed the shrinkage. In my simulation, the gradient increased by approximately 30% compared to the original design.
The elimination of sand casting defects also reduced the need for costly repairs and rework. The production efficiency improved, and the cost per casting decreased significantly. This case demonstrates that a systematic approach combining numerical simulation, process parameter optimization, and careful venting design can successfully overcome the challenges posed by complex stainless steel castings prone to sand casting defects. I now apply these principles to other similar box and valve body castings, consistently achieving high quality.
In conclusion, the key to eliminating sand casting defects in this box casting lay in understanding the specific material behavior of 06Cr13Ni4Mo, using simulation to guide riser and chill placement, and ensuring adequate venting. The results validated that even with extremely demanding requirements, defects such as cold shuts, blowholes, and shrinkage can be completely avoided, leading to a robust and reliable casting process.