I work in a foundry that uses self-hardening furan resin sand for the production of wind power components and injection molding machine components. One of our bearing housing castings, a ductile iron casting that uses a slant neck riser, began to show white spots at the outer edge of the riser neck. The white spots were located between the casting surface and the coating layer. Based on the depth reached during grinding, I found that the deepest white spot was approximately 2 mm below the casting surface. Shot blasting did not remove the condition. Additional grinding was required, which increased grinding man-hours, raised casting cost, and in some cases threatened the customer delivery schedule. This study was performed to identify the cause, reproduce the phenomenon, analyze the white substance, and implement a durable process solution for ductile iron casting production.
1. Problem statement and initial observations
I first separated the white spot from common sand-related defects. The particle morphology on the casting surface did not resemble loose sand, burned sand, or mechanical sand inclusion. The same location on the coating layer also showed a similar white appearance. Because the molding process, melting process, pouring process, and operator actions showed no unusual deviation, I focused on the interface among liquid iron, coating, and sand mold. The slant neck riser provides a local high-temperature field at the outer edge of the riser neck. In the range of 1200 °C to 1400 °C, a reaction can occur at the surface of the ductile iron casting. This reaction can create a silicon-rich layer, which appears as a white spot.
| Observation item | Condition in the ductile iron casting | Initial interpretation |
|---|---|---|
| Defect name | White spot at slant neck riser edge | Surface reaction layer, not loose sand |
| Location | Outer edge of riser neck | Local thermal condition from slant neck riser |
| Position in section | Between casting surface and coating layer | Interfacial reaction between iron and coating |
| Maximum depth | About 2 mm below casting surface | Penetrating reaction or diffusion zone |
| Shot blasting response | Not removed | Strong bonding, not a loose deposit |
| Production impact | Extra grinding, higher cost, delivery risk | Requires process control, not only cleaning |
The term ductile iron casting is important here because the metallurgical condition differs from gray iron or steel. In ductile iron casting, magnesium, cerium or other rare earth elements, carbon, silicon, aluminum, and hydrogen can participate in interfacial reactions. When a slant neck riser is used, the riser neck remains hot for a longer time. This extended hot period increases the opportunity for silica-bearing coating components to react and for silicon-rich products to form on the surface of the ductile iron casting.
2. Thermal and chemical hypothesis
I treated the white spot as a local thermochemical phenomenon. The slant neck riser changes the local modulus and solidification time. The outer edge of the riser neck receives heat from both the riser body and the casting section. This creates a thermal gradient toward the mold wall. The coating acts as a resistance layer between the liquid iron and the sand mold.
The heat content of the liquid iron can be written as:
$$Q = m c_p \Delta T = \rho V c_p (T_p – T_0)$$
where \(Q\) is heat, \(m\) is mass, \(c_p\) is specific heat, \(\Delta T\) is superheat or temperature change, \(\rho\) is density, \(V\) is volume, \(T_p\) is pouring temperature, and \(T_0\) is initial mold temperature.
The local solidification time is often related to the casting modulus:
$$t_f = B M^n$$
where \(t_f\) is freezing time, \(B\) is a constant for the alloy and mold, \(M\) is the modulus, and \(n\) is an exponent commonly close to 2 for many casting conditions. The modulus is:
$$M = \frac{V}{A}$$
For the riser and riser neck, I used the following relationships:
$$M_r = \frac{V_r}{A_r}$$
$$M_n = \frac{V_n}{A_n}$$
where \(M_r\) is the riser modulus, \(M_n\) is the neck modulus, \(V_r\) is riser volume, \(A_r\) is riser surface area, \(V_n\) is neck volume, and \(A_n\) is neck surface area. A slant neck riser can produce a hot spot at the outer edge of the neck, because the neck geometry concentrates heat and delays complete solidification. This hot spot is one of the primary conditions for white spot formation in the ductile iron casting.
The heat flux through the coating and mold can be approximated by:
$$q = -k \nabla T$$
where \(q\) is heat flux, \(k\) is thermal conductivity, and \(\nabla T\) is the temperature gradient. The coating thermal resistance is:
$$R_c = \frac{L_c}{k_c A_c}$$
where \(R_c\) is coating thermal resistance, \(L_c\) is coating thickness, \(k_c\) is coating thermal conductivity, and \(A_c\) is coating area. A thin, discontinuous, or low-refractoriness coating cannot provide enough resistance and isolation. As a result, the reaction between the iron and the silica-bearing coating becomes more intense.
The main chemical reactions proposed for the white spot formation are:
$$\mathrm{SiO_2 + (*) \rightarrow SiO_{gas}}$$
where \((*)\) represents elements or species available in ductile iron casting, such as Mg, C, Si, Al, and H. The gaseous silicon monoxide can then react further:
$$2\mathrm{SiO_{gas}} + \mathrm{O_2} \rightarrow 2\mathrm{SiO_2}$$
$$2\mathrm{SiO_{gas}} \rightarrow \mathrm{SiO_2 + Si}$$
The thermodynamic driving force can be expressed as:
$$\Delta G = \Delta H – T\Delta S$$
At high temperature, the Gibbs free energy change becomes favorable for the formation of gaseous SiO and for the subsequent deposition of silicon-rich species. The equilibrium constant can be written as:
$$K = \frac{p_{SiO}}{a_{(*)}}$$
where \(p_{SiO}\) is the partial pressure of silicon monoxide and \(a_{(*)}\) is the activity of the reactive species in the liquid iron. When the local temperature is high and the coating contains a large amount of silica, the partial pressure of SiO can increase, and the white silicon-rich layer can form on the surface of the ductile iron casting.
| Thermal factor | Effect at slant neck riser edge | Consequence for white spot |
|---|---|---|
| Local modulus | Neck edge remains hot longer | Extends reaction time |
| Pouring temperature | 1340 °C to 1350 °C in trial | Provides superheat for SiO formation |
| Riser geometry | Heat concentration at outer edge | Creates local 1200 °C to 1400 °C zone |
| Coating thickness | Thin spots allow metal penetration | Increases interfacial reaction |
| Coating composition | High silica content supplies Si and O | Promotes silicon-rich layer |
| Mold moisture | Moisture can increase oxidizing conditions | May accelerate gas reactions |
3. Experimental reproduction of the white spot
Because the white spot could not be sampled directly from a production casting without disturbing the surface, I designed a trial to reproduce the condition. The trial used the same general wall thickness, molding condition, pouring temperature, and iron chemistry as the production ductile iron casting. A test block with a slant neck riser was molded, poured, cleaned, and sectioned at the white spot location.
| Trial item | Description |
|---|---|
| Molding sand | Self-hardening furan resin sand |
| Coating mass | 20 kg of standard foundry coating |
| Riser | One slant neck riser, 150 type |
| Test block mold | 600 mm × 300 mm × 75 mm |
| Ceramic tube | 40 mm diameter, several pieces |
| Flask set | 700 mm × 700 mm × 200 mm and 700 mm × 700 mm × 300 mm |
| Pouring temperature | 1340 °C to 1350 °C |
| Coating | Standard coating applied to the test block mold |
The trial arrangement placed the slant neck riser at the side of the test block. A ceramic tube connected the bottom of the riser to the test block. After molding, coating, and core setting, the mold was closed and poured. After shakeout, the test block and riser were cleaned. The white spot region was sectioned, mounted, and prepared for scanning electron microscopy and energy dispersive spectroscopy.
| Process step | Control in trial | Purpose |
|---|---|---|
| Riser selection | Slant neck riser | Reproduce hot spot at riser neck edge |
| Mold assembly | Riser on side of block | Match production thermal geometry |
| Connection | Ceramic tube from bottom | Control filling path |
| Sand filling | Furan resin sand | Match production mold medium |
| Pattern removal | Standard strip and repair | Maintain mold surface quality |
| Coating | Standard coating sequence | Evaluate coating effect |
| Pouring | 1340 °C to 1350 °C | Match production superheat |
| Cleaning | Shakeout and blast | Reveal white spot |
4. Iron chemistry and treatment control
The trial iron chemistry was controlled to match the production ductile iron casting. The carbon and silicon contents were kept within a narrow range. Magnesium and rare earth contents were controlled to achieve good nodularity. Sulfur after treatment was kept low. The purpose was not to change the metallurgy, but to reproduce the white spot at the slant neck riser edge.
| Element or parameter | Control requirement |
|---|---|
| Carbon, original iron | 3.40% to 3.45% |
| Silicon, original iron | 2.83% to 2.93% |
| Silicon, casting | 3.50% to 3.60% |
| Manganese | Less than 0.025% |
| Phosphorus | 0.040% maximum |
| Sulfur, before treatment | 0.025% maximum |
| Sulfur, after treatment | 0.005% to 0.015% |
| Magnesium | 0.035% to 0.055% |
| Rare earth | Less than 0.010% |
| Carbon equivalent | 4.56% to 4.65% |
| Antimony addition | 0.005% |
| Nickel addition | Not used in this trial |
| Treatment stage | Type | Ratio or addition | Purpose |
|---|---|---|---|
| Nodularizer component A | Magnesium-bearing alloy | 70% of total | Control nodularization |
| Nodularizer component B | Rare earth bearing alloy | 30% of total | Balance nodularization |
| Total nodularizer | Sandwich or tundish method | 0.90% to 1.20% | Produce ductile iron casting |
| Inoculant | BS-1A type | Pocket addition 0.10% to 0.70% | Control graphite shape |
| Melting route | Induction furnace or cupola plus holding furnace | As scheduled | Maintain chemistry |
| Test block type | Cast-on test block | As required | Verify structure |
In my experience, the ductile iron casting chemistry itself did not directly create the white spot. Instead, the chemistry provided reactive species at the interface. Magnesium, carbon, silicon, aluminum, and hydrogen can all participate in the high-temperature reaction with silica-bearing coating. The slant neck riser then provides the necessary temperature and time for the reaction to proceed.
5. Sampling and analytical methods
I selected the white spot region from the sectioned test block. The sample was examined by scanning electron microscopy at low magnification, then by elemental mapping and point analysis. The purpose was to determine whether the white substance was sand, coating residue, slag, or a reaction product.
| Method | Purpose | Observation target |
|---|---|---|
| Visual inspection | Confirm location and morphology | White spot at riser neck edge |
| Sectioning | Expose depth profile | Layer between casting and coating |
| Scanning electron microscopy | Observe microscopic morphology | Surface layer and penetration zone |
| Elemental mapping | Map Si, O, Fe, Ti, Mn, S | Distribution of white substance |
| Point analysis | Measure local composition | Representative points 1 to 4 |
| Depth measurement | Estimate layer thickness | Tens of μm to 2 mm |
The sample surface contained white material. At 25× magnification, the white substance appeared as a distributed layer. The elemental maps showed that silicon and oxygen were concentrated in the white area. Iron was distributed over the rest of the field, except where silicon was concentrated. Titanium was mostly associated with marker paint and was not considered part of the defect. Sulfur, manganese, and other minor elements appeared only in small amounts.
| Element | Distribution in elemental map | Interpretation |
|---|---|---|
| Si | Strongly concentrated in white area | Main component of white spot |
| O | Overlaps Si distribution | Silicon oxide or oxidized layer |
| Fe | Present outside Si-rich zones and in dark areas | Metal matrix and iron penetration |
| Ti | Minor, mostly from marker paint | Not a major cause |
| S | Trace amount | Minor impurity or residual |
| Mn | Trace amount | Minor alloying element |
6. Energy dispersive spectroscopy results
I selected four representative points for energy dispersive spectroscopy. The following tables summarize the measured mass fraction and atomic fraction. The results consistently show that Si and O are the dominant elements in the white substance. The dark areas are iron-rich, which indicates that liquid iron penetrated into the coating or mold surface. This penetration supports the conclusion that the white spot is not simply a coating residue, but a reaction product formed at the interface of the ductile iron casting and the coating.
| Element | Mass fraction at point 1, % | Atomic fraction at point 1, % |
|---|---|---|
| C | 23.46 | 34.07 |
| O | 46.92 | 51.17 |
| Si | 17.57 | 10.92 |
| S | 0.35 | 0.19 |
| Fe | 11.70 | 3.66 |
| Element | Mass fraction at point 2, % | Atomic fraction at point 2, % |
|---|---|---|
| C | 10.81 | 17.45 |
| O | 45.94 | 55.66 |
| Si | 34.40 | 23.74 |
| S | 0.24 | 0.14 |
| Ti | 0.18 | 0.07 |
| Mn | 0.34 | 0.12 |
| Fe | 8.09 | 2.81 |
| Element | Mass fraction at point 3, % | Atomic fraction at point 3, % |
|---|---|---|
| C | 11.27 | 17.73 |
| O | 47.59 | 56.22 |
| Si | 36.15 | 24.33 |
| Ti | 0.55 | 0.22 |
| Fe | 4.43 | 1.50 |
| Element | Mass fraction at point 4, % | Atomic fraction at point 4, % |
|---|---|---|
| C | 19.56 | 29.69 |
| O | 42.27 | 48.17 |
| Si | 29.75 | 19.31 |
| S | 0.45 | 0.26 |
| Ti | 0.23 | 0.09 |
| Mn | 0.20 | 0.07 |
| Fe | 7.29 | 2.38 |
| In | 0.26 | 0.04 |
The point analysis results can be summarized as follows. The white substance is mainly silicon oxide. The silicon content in the white area is significantly higher than the nominal silicon content of the ductile iron casting. The oxygen content is also high, which confirms oxidation. Iron is present but is lower in the white area than in the dark area. This supports the conclusion that a silicon-rich oxide layer formed on the casting surface and that iron penetration occurred in adjacent areas.
| Point | Dominant elements | Minor elements | Interpretation |
|---|---|---|---|
| 1 | O, C, Si, Fe | S | Mixed oxide and metal region |
| 2 | O, Si, C, Fe | S, Ti, Mn | Strong silicon oxide layer |
| 3 | O, Si, C | Ti, Fe | High silicon oxide region |
| 4 | O, Si, C, Fe | S, Ti, Mn, In | Silicon-rich layer with iron penetration |
7. Interpretation of the white spot mechanism
Based on the scanning electron microscopy and energy dispersive spectroscopy results, I concluded that the white spot is a silicon-rich layer formed by a high-temperature reaction between the silica-bearing coating and the ductile iron casting. The slant neck riser creates a local hot spot at the outer edge of the riser neck. This hot spot raises the local temperature to the range where silica can be reduced or decomposed into gaseous SiO. The gaseous SiO can then oxidize again or decompose to form SiO₂ and Si. The resulting silicon-rich layer appears white after cleaning.
The reaction sequence can be summarized as:
$$\mathrm{SiO_2 + (*) \rightarrow SiO_{gas}}$$
$$2\mathrm{SiO_{gas}} + \mathrm{O_2} \rightarrow 2\mathrm{SiO_2}$$
$$2\mathrm{SiO_{gas}} \rightarrow \mathrm{SiO_2 + Si}$$
The diffusion of silicon into the surface of the ductile iron casting can be approximated by Fick’s second law:
$$\frac{\partial C}{\partial t} = D \frac{\partial^2 C}{\partial x^2}$$
For a simple estimate of diffusion distance, I used:
$$\delta = \sqrt{D t}$$
where \(\delta\) is diffusion distance, \(D\) is diffusion coefficient, and \(t\) is time. At high temperature, both \(D\) and \(t\) increase, so the silicon-rich layer can grow to a depth of tens of micrometers or even up to about 2 mm. This explains why shot blasting cannot remove the white spot completely. The layer is not a loose surface deposit. It is chemically bonded or metallurgically integrated into the surface of the ductile iron casting.
| Mechanism step | Condition | Result |
|---|---|---|
| Local heating | Slant neck riser edge remains hot | Temperature reaches 1200 °C to 1400 °C |
| Silica reaction | Coating contains SiO₂ | Formation of gaseous SiO |
| Gas transport | Gas moves through coating and mold interface | Silicon species reach casting surface |
| Deposition | Oxidation or decomposition of SiO | SiO₂ and Si deposit on surface |
| Diffusion | High temperature and long time | Silicon-rich layer grows inward |
| Cleaning resistance | Layer is bonded to surface | Shot blasting is insufficient |
The dark area observed in the sample is also important. The iron-rich dark area indicates that liquid iron penetrated into the coating layer. This penetration can occur when the coating is too thin, when the coating is not well sintered, or when the coating has low resistance to high-temperature erosion. Once liquid iron penetrates the coating, the contact area between iron and silica-bearing material increases. The reaction then becomes more severe, and the white spot becomes more likely to form on the ductile iron casting.
8. Improvement measures
After confirming the cause, I focused on the coating material and the coating operation. The goal was to increase the refractoriness of the coating, reduce the amount of free silica available for reaction, improve the coating penetration and layer thickness, and ensure that the coating remains stable during the entire pouring and solidification period. The improvements were implemented in two areas: coating formulation and operator practice.
8.1 Coating material improvement
I increased the zircon flour content in the coating aggregate to 30%. Zircon flour has high refractoriness and can reduce the amount of silica available at the interface. This change improves the high-temperature stability of the coating and helps the coating form a sintered layer that isolates the liquid iron from the mold sand. I also optimized the aggregate particle size distribution so that the coating can penetrate 3 mm to 5 mm into the mold surface and form a protective layer. The coating thickness on the mold surface was controlled to 0.35 mm to 0.5 mm.
| Coating parameter | Before improvement | After improvement | Expected effect |
|---|---|---|---|
| Zircon flour content | Lower, silica-rich aggregate | 30% zircon flour | Increase refractoriness and reduce Si reaction |
| Aggregate particle size | Broad and not optimized | Optimized for penetration and sintering | Improve coating density and isolation |
| Penetration layer | Inconsistent | 3 mm to 5 mm | Protect sand mold surface |
| Coating thickness | Variable, sometimes thin | 0.35 mm to 0.5 mm | Provide reliable shield |
| High-temperature stability | Limited | Improved | Resist metal penetration |
| Sintering behavior | Uneven | Controlled | Facilitate cleaning and peeling |
The coating layer thickness can be estimated from the applied mass, density, and area:
$$t_c = \frac{m_c}{\rho_c A}$$
where \(t_c\) is coating thickness, \(m_c\) is coating mass, \(\rho_c\) is coating density, and \(A\) is coated area. This relationship helped me standardize the coating amount for the riser neck region. It also helped the operators understand that a thicker coating is not always better. An excessively thick coating can crack, spall, or trap gas. The target is a dense, uniform, well-sintered layer that remains stable at high temperature.
8.2 Coating application improvement
I specified the coating Baume degree for each application pass. The first pass was controlled at 38 °Be to 40 °Be. The second pass was controlled at 55 °Be to 60 °Be. The third pass was controlled at 45 °Be to 50 °Be. The Baume degree is related to density and solids content. A controlled Baume degree helps maintain the correct coating viscosity, solids content, and penetration behavior. For a liquid heavier than water, the approximate relationship is:
$$\rho = \frac{145}{145 – ^\circ Be}$$
where \(\rho\) is specific gravity and \(^\circ Be\) is the Baume reading. This formula was used as a practical guide for coating preparation and dilution. The operators were trained to check the Baume degree before each coating pass and to record the result.
| Coating pass | Baume control | Purpose |
|---|---|---|
| First pass | 38 °Be to 40 °Be | Promote penetration into mold surface |
| Second pass | 55 °Be to 60 °Be | Build coating thickness and strength |
| Third pass | 45 °Be to 50 °Be | Level and seal the coating surface |
| Riser neck edge | Special attention | Prevent thin spots and metal penetration |
| Drying | Blowtorch baking | Remove moisture and strengthen coating |
After coating, the outer edge of the riser neck was baked with a blowtorch. This step removes residual moisture and improves the coating strength. Moisture can react with liquid iron and produce oxidizing conditions. It can also reduce coating adhesion and create gas defects. The blowtorch baking was made mandatory for the region where the white spot had occurred. The operators were also trained to apply additional coating attention to the riser neck edge, because this area is exposed to the highest temperature and the longest solidification time.
| Operator control | Requirement | Reason |
|---|---|---|
| Baume check | Record each pass | Maintain coating viscosity and solids |
| Brush control | Uniform strokes, no thin spots | Avoid local low thickness |
| Riser neck edge | Extra coating and smoothing | Resist high-temperature hot spot |
| Blowtorch baking | Bake after coating | Remove moisture and strengthen layer |
| Hot air time | Follow specified schedule | Ensure mold cavity dryness |
| Moisture check | Use temperature and humidity meter | Confirm sand and mold dryness |
8.3 Mold and sand control
I also reviewed the mold and sand condition. Moisture in the furan resin sand can increase the oxidizing potential at the interface. If the mold cavity is not dry, water vapor can react with the liquid iron and produce additional oxygen. This oxygen can participate in the SiO oxidation reaction. For this reason, I specified a hot air time and checked the internal temperature and humidity of the sand mold. The mold cavity was kept dry before pouring. The sand mixing, resin addition, and curing time were also checked to ensure consistent mold strength and surface stability.
| Sand and mold variable | Control direction | Effect on white spot |
|---|---|---|
| Sand moisture | Keep low and stable | Reduce oxidizing gas |
| Resin addition | Within specification | Maintain mold surface strength |
| Curing time | Sufficient for full cure | Prevent mold surface erosion |
| Mold cavity dryness | Hot air and humidity check | Reduce gas reaction |
| Coating drying | Blowtorch and air dry | Improve coating strength |
| Pouring temperature | Controlled within range | Avoid excessive superheat |
9. Verification and results
After implementing the coating material change, the Baume control, the blowtorch baking, and the operator training, I tracked the slant neck riser region on the ductile iron casting. The white spot condition was significantly reduced. In the initial verification lots, the white spot was either eliminated or limited to a very shallow surface condition that could be removed by normal shot blasting. The additional grinding operation was no longer required in most cases. The cleaning time and grinding cost decreased. The delivery risk caused by extra surface repair was also reduced.
| Verification item | Before improvement | After improvement | Result |
|---|---|---|---|
| White spot at riser neck edge | Frequent and deep | Rare and shallow | Significant reduction |
| Maximum depth | About 2 mm | Not measurable or very shallow | Improved |
| Shot blasting removal | Not removed | Mostly removable | Improved |
| Extra grinding | Required | Usually not required | Cost reduced |
| Coating layer quality | Uneven, thin spots | Uniform, 0.35 mm to 0.5 mm | Better isolation |
| Coating penetration | Inconsistent | 3 mm to 5 mm | Better mold protection |
| Operator compliance | Variable | Standardized | More repeatable |
| Delivery impact | Schedule risk | Controlled | Improved |
The improvement was not based on a single change. It was the combined result of increasing zircon flour content, reducing silica exposure, optimizing aggregate size, controlling coating Baume degree, maintaining coating thickness, baking the riser neck edge, and keeping the mold cavity dry. The slant neck riser still provides a local hot spot, but the improved coating now resists the high-temperature reaction long enough to prevent the white spot from forming on the ductile iron casting.
| Improvement action | Technical reason | Observed effect |
|---|---|---|
| Increase zircon flour to 30% | Higher refractoriness, lower free silica | Less silicon-rich reaction |
| Optimize aggregate size | Better packing and penetration | Denser coating layer |
| Control penetration layer 3 mm to 5 mm | Protect sand surface | Less metal penetration |
| Control coating thickness 0.35 mm to 0.5 mm | Balance insulation and strength | Better shield effect |
| Control Baume per pass | Stable viscosity and solids | Repeatable coating quality |
| Blowtorch bake riser neck edge | Remove moisture and strengthen coating | Fewer interfacial reactions |
| Train operators and track process | Reduce human variation | Consistent application |
| Control mold dryness | Reduce oxidizing gases | Less white spot tendency |
10. Process control plan for future production
To sustain the improvement, I created a control plan for the ductile iron casting with the slant neck riser. The control plan focuses on the riser neck region, coating material, coating application, mold dryness, and pouring temperature. The most critical control point is the outer edge of the riser neck, because this is where the slant neck riser creates the highest local temperature and the longest reaction time.
| Control point | Specification | Frequency | Responsibility |
|---|---|---|---|
| Zircon flour content | 30% in coating aggregate | Each coating batch | Coating preparation |
| Coating Baume, first pass | 38 to 40 °Be | Each mold | Molding operator |
| Coating Baume, second pass | 55 to 60 °Be | Each mold | Molding operator |
| Coating Baume, third pass | 45 to 50 °Be | Each mold | Molding operator |
| Coating thickness | 0.35 mm to 0.5 mm | Each mold for critical area | Molding operator |
| Coating penetration | 3 mm to 5 mm | Process audit | Process engineer |
| Riser neck edge baking | Blowtorch until dry | Each mold | Molding operator |
| Mold cavity dryness | No visible moisture | Each mold | Molding operator |
| Sand temperature and humidity | Within specified range | Each shift | Sand control |
| Pouring temperature | 1340 °C to 1350 °C for trial; production range as specified | Each pour | Melting operator |
| Visual inspection after cleaning | No white spot requiring extra grinding | Each casting | Cleaning and inspection |
The control plan also includes a reaction plan. If a white spot appears again, the operator checks the coating Baume degree, coating thickness, blowtorch baking record, and mold dryness. If the white spot is deep, the process engineer reviews the coating batch, zircon flour content, aggregate size distribution, and pouring temperature. If the white spot is shallow, the cleaning team increases shot blasting time and verifies whether the condition is only a surface staining. This structured response prevents the defect from returning to the previous level.
| Abnormal condition | Immediate check | Corrective action |
|---|---|---|
| White spot at riser neck edge | Coating thickness and baking record | Re-coat and bake critical area |
| Coating Baume out of range | Density and solids content | Adjust dilution and mix |
| Thin coating spot | Brush marks and coverage | Add coating pass and smooth |
| Mold moisture | Humidity meter reading | Increase hot air time |
| Excessive pouring temperature | Thermocouple record | Adjust pouring practice |
| Deep white spot | Coating batch and aggregate | Verify zircon content and particle size |
11. Discussion of key variables
The white spot problem in the ductile iron casting is controlled by the interaction of several variables. I summarize the most important variables and their effects below. The most influential variables are local temperature, coating composition, coating thickness, and coating stability at high temperature. The slant neck riser geometry is not easily removed because it is required for feeding. Therefore, the process solution must focus on the coating and the interfacial conditions.
| Variable | Low condition | High condition | Effect on white spot |
|---|---|---|---|
| Local temperature | Fast cooling, short reaction time | Long hot spot, extended reaction | High temperature promotes SiO formation |
| Silica in coating | Low silica, high zircon | High silica, low zircon | High silica increases white spot risk |
| Coating thickness | Thin, weak shield | Thick, possibly cracked | Optimum 0.35 mm to 0.5 mm |
| Coating penetration | Shallow, weak mold protection | Deep, strong mold protection | Optimum 3 mm to 5 mm |
| Baume degree | Too thin or too thick | Out of specification | Controlled per pass |
| Mold moisture | Dry mold | Wet mold | Moisture increases oxidation |
| Pouring temperature | Low superheat | High superheat | High superheat increases reaction |
| Riser neck geometry | Small hot spot | Large hot spot | Slant neck creates local hot spot |
The coating can be viewed as a thermal and chemical barrier. Its thermal resistance is proportional to thickness and inversely proportional to thermal conductivity. Its chemical resistance depends on the refractoriness of the aggregate, the binder system, and the sintering behavior. When the coating is optimized, it can slow the heat transfer and isolate the silica-bearing mold sand from the liquid iron. This reduces the amount of SiO gas that can reach the surface of the ductile iron casting. As a result, the white spot is prevented.
The coating penetration layer is also important. The Washburn equation can be used as a practical description of coating penetration into the mold surface:
$$L^2 = \frac{\gamma r \cos\theta}{2\mu} t$$
where \(L\) is penetration distance, \(\gamma\) is surface tension, \(r\) is capillary radius, \(\theta\) is contact angle, \(\mu\) is viscosity, and \(t\) is time. By controlling Baume degree, viscosity, and aggregate size, I can influence the penetration distance. A penetration layer of 3 mm to 5 mm provides a protective zone in the mold surface. This zone reduces metal penetration and protects the sand from direct contact with the liquid iron.
12. Practical lessons for ductile iron casting production
This investigation provided several practical lessons that I now apply to other ductile iron casting jobs with slant neck risers or similar hot spots. The first lesson is that a white spot at a riser neck edge should not be treated as a simple cleaning problem. It is usually an interfacial reaction that requires process control. The second lesson is that coating composition and coating application are both important. A good coating material can still fail if it is applied at the wrong Baume degree, at the wrong thickness, or without proper drying. The third lesson is that the riser neck outer edge is a critical zone. It deserves extra coating, extra drying, and extra inspection.
| Lesson | Explanation | Action in production |
|---|---|---|
| White spot is not loose sand | It is a bonded silicon-rich layer | Use process control, not only shot blasting |
| Slant neck riser creates hot spot | Local temperature and time increase | Focus on riser neck outer edge |
| Coating composition matters | Silica can react with iron | Increase zircon flour and reduce free silica |
| Coating application matters | Thin spots and moisture cause defects | Control Baume, thickness, and baking |
| Mold dryness matters | Moisture increases oxidation | Hot air and humidity check |
| Operator training matters | Human variation causes recurrence | Standard work instruction and audit |
| Verification matters | Improvement must be measured | Track white spot rate and grinding time |
I also learned that the white spot can be present at different depths. In severe cases, it can extend about 2 mm below the surface. In mild cases, it may be a thin surface layer. The depth depends on the local temperature, the reaction time, and the coating condition. For this reason, the inspection method should include visual inspection after shot blasting, plus grinding checks at the riser neck edge when a new process or new coating batch is introduced. This helps detect the defect before it reaches the customer.
| Depth condition | Appearance | Typical cause | Response |
|---|---|---|---|
| Thin surface layer | Light white stain | Short reaction or mild coating reaction | Shot blasting and coating adjustment |
| Moderate layer | Visible white spot after blasting | Thin coating or high local temperature | Increase coating and baking |
| Deep layer | White spot after grinding, up to 2 mm | Severe SiO reaction and metal penetration | Change coating composition and process |
13. Conclusion
The white spot at the outer edge of the slant neck riser in the ductile iron casting was caused by a high-temperature reaction between the silica-bearing coating and the casting surface. The slant neck riser provided the local temperature field between 1200 °C and 1400 °C, which promoted the formation of gaseous SiO and the subsequent deposition of a silicon-rich layer. Energy dispersive spectroscopy confirmed that the white substance was mainly Si and O, with iron in adjacent dark areas. The white spot was not loose sand and could not be removed by shot blasting alone. Its depth could reach about 2 mm.
I solved the problem by increasing the zircon flour content in the coating to 30%, optimizing the aggregate particle size, controlling the coating penetration layer to 3 mm to 5 mm, controlling the coating thickness to 0.35 mm to 0.5 mm, specifying the Baume degree for each coating pass, baking the riser neck edge with a blowtorch, and controlling mold dryness. These measures improved the high-temperature stability of the coating, reduced the amount of free silica available for reaction, and increased the resistance of the coating to metal penetration. The white spot condition was effectively controlled, and the extra grinding cost and delivery risk were reduced.
| Conclusion item | Finding | Production action |
|---|---|---|
| Defect identity | Silicon-rich oxide layer | Treat as interfacial reaction |
| Thermal condition | Slant neck riser hot spot | Focus on riser neck outer edge |
| Chemical condition | SiO formation and deposition | Reduce silica and increase zircon |
| Coating formulation | Zircon flour 30% | Improve refractoriness |
| Coating application | Baume and thickness control | Standardize each pass |
| Drying | Blowtorch and hot air | Remove moisture and strengthen layer |
| Result | White spot controlled | Lower cleaning cost and delivery risk |
For future ductile iron casting production, I will continue to treat the slant neck riser edge as a critical control zone. I will monitor coating Baume degree, coating thickness, penetration depth, blowtorch baking, mold dryness, and pouring temperature. I will also continue to train operators and audit the coating process. The white spot problem demonstrated that a small interfacial reaction can create a large production cost. By understanding the thermal and chemical mechanism, and by controlling the coating system, the ductile iron casting process can be made more stable and more cost-effective.