The 00BOB cylinder is a critical high-temperature and high-pressure cast steel component in supercritical steam turbines. It seals the flow path from the atmosphere, forming an enclosed steam chamber to ensure energy conversion within the turbine. This product is made from ZG25CrMoV-1+7 cast steel. Improper gating system design and pouring operations can easily lead to sand holes, gas porosity, and cracks. In our production, we encountered severe sand casting defects such as core sand adhesion, gas porosity, and surface cracking near the midline flow baffle region. The defect area exhibited clogging of the baffle holes and extensive sand inclusion.
The affected zone required manual cleaning, revealing deep sand penetration and localized shrinkage cavities. To systematically resolve these sand casting defects, we conducted a thorough process investigation and root cause analysis.
Root Cause Analysis of Sand Casting Defects
Process Investigation
Two 00BOB cylinders were molded and cored, then closed and poured after two days. During this period, the workshop ambient temperature was 10–15°C with a relative humidity of 9–11%. The furan resin sand system had a compressive strength of 8.5 MPa. After closing the mold, negative pressure treatment was applied to remove loose sand and debris. The molten steel cleanliness met the requirements of JB/T10087-2001 for steam turbine pressure-bearing steel castings. However, during pouring, both castings exhibited riser boiling phenomena. Post-machining UT inspection near the riser zone at depths of 60–100 mm revealed sand casting defects including sand adhesion, gas porosity, and surface micro-cracks.
Defect Mechanism Analysis
Core Floating: As shown in our assembly design, the entire sand core was fixed and located solely by the core print at one end. The core tip that defined the baffle channel was unsupported and located exactly at the mold parting line. The riser was placed at the top of the core print. During pouring, the erosive flow of molten steel generated buoyancy forces that overcame the core print fixation, causing core floating. This misalignment between the sand mold and core blocked the riser channel, preventing slag from rising into the riser — resulting in slag inclusion. The narrowed riser path also restricted venting, leading to boiling and intense gas evolution. The trapped gas could not escape promptly, forming dense gas porosity in the surrounding matrix. Additionally, the riser lost its feeding capability, inducing shrinkage porosity and cavities.
Poor Permeability: Table 1 summarizes the physical properties of the reclaimed sand used for this cylinder casting. The data indicate that the sand grain size was too fine, the loss on ignition was abnormally high, and the micro-powder content exceeded acceptable limits. These factors significantly reduced the mold permeability. Consequently, gases generated during pouring could not be evacuated efficiently, leading to extensive gas porosity in the casting body.
| Sieve Mesh (mesh) | Loss on Ignition (%) | Residue after Fine Powder (g) | Residue before Regeneration (g) | Residue after Regeneration (g) |
|---|---|---|---|---|
| 28 | 3.1 | 3.2 | 2.9 | – |
| 45 | 31.2 | 31.3 | 29.9 | – |
| 55 | 30.3 | 29.7 | 29.3 | – |
| 75 | 23.2 | 21.1 | 21.3 | – |
| 100 | 10.2 | 3.2 | 0.7 | – |
| 150 | 10.4 | 2.9 | 0.8 | – |
| 260 | 9.2 | 4.3 | 0.7 | – |
The fine grains and high micro-powder content (e.g., residues at 100 and 260 mesh) severely reduced the mold’s gas permeability. Using the permeability equation:
$$ K = \frac{V \cdot H}{A \cdot p \cdot t} $$
where \(K\) is the permeability coefficient, \(V\) is the volume of gas passed, \(H\) is the sample height, \(A\) is the cross-sectional area, \(p\) is the pressure difference, and \(t\) is the time. With finer sand, the effective pore channels become narrower, drastically decreasing \(K\) and increasing gas backpressure, which contributes directly to gas porosity formation — one of the most common sand casting defects.
Poor Collapsibility: Furan resin sand exhibits high rigidity and a large thermal expansion coefficient, leading to insufficient collapsibility. The CrMoV steel grade itself is prone to hot cracking. The complex geometry of the 00BOB cylinder with large wall-thickness variations, combined with the inadequate collapsibility of the internal sand core, obstructed the free contraction of the casting during solidification. Sharp corners in the mold experienced intense local heating, generating thermal stresses that exceeded the material’s strength, resulting in the formation of surface micro-cracks. The thermal strain can be expressed as:
$$ \epsilon = \alpha \cdot \Delta T $$
where \(\epsilon\) is the thermal strain, \(\alpha\) is the linear expansion coefficient, and \(\Delta T\) is the temperature change. When the core cannot deform, this strain translates into tensile stress, promoting crack initiation — another critical sand casting defect.
Preventive Measures for Sand Casting Defects
Based on our analysis, we implemented a series of corrective actions to eliminate these sand casting defects:
- Prevention of sand adhesion: We relocated the riser from the core tip area to the central region of the casting, away from the unsupported core end. Between the risers at the parting line, we added slag collection pockets to facilitate the floatation of slag and impurities. Sufficient molten steel volume was ensured to avoid shrinkage porosity. Moreover, we redesigned the core support system by adding a rigid support to fix the cantilevered core tip, as illustrated in our schematic.
- Prevention of gas porosity: We tightened control over reclaimed sand sieving, ensuring that the residue on the 100-mesh sieve did not exceed 10% and that the micro-powder content was below 2%. Additional vent holes (both direct and bypass) were drilled into the sand core and mold. Gas-permeable ropes were embedded in the mold to reduce flow turbulence and prevent gas entrapment.
- Prevention of cracks: To improve collapsibility, we filled the hollow interior of the sand core with dry sand and foam. The external chill surfaces were polished smooth and thoroughly dried to avoid chilling-induced cracks.
Results and Conclusion
After implementing these changes, we produced six additional 00BOB cylinder castings, all of which passed rigorous inspection without any sand casting defects. Practical experience confirms that enhancing the stability and collapsibility of the sand core, controlling the screening and micro-powder content of reclaimed sand, and increasing the mold permeability are highly effective in eliminating sand adhesion, gas porosity, and cracks in 00BOB cylinders. These measures significantly improved both the internal quality and surface finish of the castings, providing a robust solution for preventing sand casting defects in pressure-bearing steel castings for steam turbines.