My research focuses on the repair of surface defects in gray cast iron components, particularly those arising from sand foundry defects. Sand foundry defects, such as porosity, slag inclusion, shrinkage, and cold shuts, are common in cast iron production and often lead to costly rejections. In this study, I employed plasma spray welding with Ni25 self-fluxing alloy powder to repair these surface defects, and compared the results with manual arc welding using Z116 and Z308 electrodes. I systematically investigated the microstructure, chemical composition, and mechanical properties of the repaired layers. The primary aim was to identify the most reliable and economical method for restoring defective gray cast iron parts, while ensuring the repaired region meets the required service performance. My findings indicate that plasma spray welding is highly effective in eliminating the formation of white iron in the semi-melted zone, which is a critical issue in conventional arc welding repairs. This research provides a practical baseline for selecting appropriate repair techniques, especially when dealing with sand foundry defects in industrial applications.
1. Introduction
Gray cast iron (HT250) is widely used in machinery manufacturing, metallurgy, mining, petrochemical, and other industrial sectors due to its simple production process and low cost. However, during the casting process, sand foundry defects such as gas pores, slag inclusions, shrinkage cavities, and cracks are inevitably introduced. Additionally, machining errors and in-service wear can create surface damages that require repair. Repairing these surface defects is of significant economic importance, as it extends the service life of components and reduces material waste. Various surface engineering technologies have been employed for cast iron repair, including oxy-acetylene flame spray welding, plasma spray welding, arc welding, laser cladding, and brazing. Among them, plasma arc spray welding stands out due to its high degree of automation, simple operation, high production efficiency, and excellent metallurgical bonding quality.
In the context of sand foundry defects, the primary challenge is to achieve a repair layer that is free from hard and brittle phases, such as white iron, which forms due to rapid cooling and the presence of carbide-stabilizing elements. Plasma spray welding, using a concentrated high-energy plasma arc, provides precise control over heat input and dilution, thereby minimizing the formation of white iron in the heat-affected zone. This study aims to compare the effectiveness of plasma spray welding with two commonly used electrodes: Z116 (high vanadium iron electrode) and Z308 (pure nickel electrode). I analyzed the microstructural evolution, phase composition, and mechanical properties of the repaired layers, including hardness, tensile strength, and crack resistance. The results demonstrate that plasma spray welding with Ni25 powder is the most suitable method for repairing HT250 components, as it produces a defect-free repair layer with mechanical properties closely matching the base material.
The occurrence of sand foundry defects is not limited to the initial casting process; they can also manifest during subsequent service as fatigue cracks or wear scars. Therefore, a robust repair technique must be able to handle both minor surface blemishes and deeper defects. Traditional arc welding methods often introduce additional problems, such as high hardness in the fusion zone and cracking, due to the uncontrolled thermal cycles. In contrast, plasma spray welding offers a lower dilution rate and a narrower heat-affected zone, making it an ideal candidate for repairing sand foundry defects. In this paper, I present a comprehensive study that includes process parameter optimization using orthogonal experiments, microstructural characterization using optical microscopy, scanning electron microscopy (SEM), energy dispersive spectroscopy (EDS), and X-ray diffraction (XRD), as well as mechanical testing. My goal is to provide a clear guideline for engineers to select the optimal repair strategy for gray cast iron components affected by sand foundry defects.
2. Materials and Methods
2.1 Base Material
I used HT250 gray cast iron plates with dimensions of 60 mm × 40 mm × 10 mm. The chemical composition of HT250 is listed in Table 1.
| C | Si | Mn | P | S |
|---|---|---|---|---|
| 3.1–3.3 | 1.6–1.8 | 0.7–0.9 | <0.25 | <0.12 |
2.2 Repair Materials
I selected Ni25 nickel-based self-fluxing alloy powder for plasma spray welding. The powder has a particle size of 50–153 mesh, an apparent density of 4.8 g/cm³, and a melting point of approximately 1300 °C. Its nominal composition is given in Table 2. For comparison, I used two types of cast iron electrodes: Z308 (pure nickel electrode with a diameter of 3.2 mm) and Z116 (high vanadium iron electrode with a diameter of 3.2 mm).
| C | Cr | B | Si | Fe | Ni |
|---|---|---|---|---|---|
| 0.05 | 5.06 | 1.58 | 3.04 | 3.2 | Balance |
2.3 Experimental Setup
The plasma spray welding system used was the EUTRONIC GAP 2001 DC from Castolin. The system consists of a controller, powder feeder, torch, and cooling unit. The arc welding was performed using an inverter DC welding machine. I prepared samples by cleaning the HT250 base metal to remove oil and rust, followed by the deposition of the repair layer. The repair layers were then sectioned into small specimens (10 mm × 10 mm × 15 mm) using wire electrical discharge machining for subsequent analysis.
The image above illustrates an automatic pouring line, which is a typical scenario where sand foundry defects can arise due to improper pouring or melt quality. The repair of such defects is critical to maintain production efficiency and reduce scrap rate.
2.4 Process Parameters
I optimized the plasma spray welding parameters using orthogonal experiments. The optimized parameters are listed in Table 3. For arc welding, I used a welding current of 100 A for both Z116 and Z308 electrodes, with preheating not applied and samples allowed to cool in air after welding.
| Parameter | Value |
|---|---|
| Transferred arc current | 55 A |
| Transferred arc voltage | 25 V |
| Plasma gas (Ar) flow rate | 1.5 L/min |
| Powder gas (Ar) flow rate | 2.0 L/min |
| Shielding gas (Ar) flow rate | 12 L/min |
| Nozzle-to-workpiece distance | 10 mm |
3. Orthogonal Experiment for Process Parameter Optimization
To determine the best combination of plasma spray welding parameters, I designed an L9(3³) orthogonal array with three factors: transferred arc current (A), plasma gas flow rate (B), and powder gas flow rate (C). Each factor was set at three levels, as shown in Table 4. The microhardness of the deposited layer was chosen as the evaluation index, because the hardness should be as close as possible to that of the base cast iron (≈238.5 HV). A hardness that is too high indicates the formation of hard phases (e.g., white iron), which are detrimental to machinability and represent a classic manifestation of unaddressed sand foundry defects.
| Level | Factor A: Current (A) | Factor B: Plasma gas (L/min) | Factor C: Powder gas (L/min) |
|---|---|---|---|
| 1 | 45 | 1.2 | 2.0 |
| 2 | 55 | 1.5 | 2.5 |
| 3 | 65 | 1.8 | 3.0 |
I conducted nine experiments according to the orthogonal array. The measured microhardness values for each specimen are summarized in Table 5. The range analysis (R) was used to determine the significance of each factor. A larger R value indicates a greater influence on the hardness.
| Specimen No. | Factor A | Factor B | Factor C | Hardness (HV) |
|---|---|---|---|---|
| 1 | 45 | 1.2 | 2.0 | 255.7 |
| 2 | 45 | 1.5 | 2.5 | 250.3 |
| 3 | 45 | 1.8 | 3.0 | 262.7 |
| 4 | 55 | 1.2 | 2.5 | 247.6 |
| 5 | 55 | 1.5 | 3.0 | 235.6 |
| 6 | 55 | 1.8 | 2.0 | 240.5 |
| 7 | 65 | 1.2 | 3.0 | 273.2 |
| 8 | 65 | 1.5 | 2.0 | 255.6 |
| 9 | 65 | 1.8 | 2.5 | 264.2 |
The average hardness for each level of each factor, denoted as \( k_{j1}, k_{j2}, k_{j3} \), was calculated. For example, for factor A:
\[
\begin{aligned}
K_{A1} &= 255.7 + 250.3 + 262.7 = 768.7 \\
k_{A1} &= 768.7 / 3 = 256.2 \\
K_{A2} &= 247.6 + 235.6 + 240.5 = 723.7 \\
k_{A2} &= 723.7 / 3 = 241.2 \\
K_{A3} &= 273.2 + 255.6 + 264.2 = 793.0 \\
k_{A3} &= 793.0 / 3 = 264.3
\end{aligned}
\]
Similarly, for factor B (plasma gas flow) and factor C (powder gas flow), the averages were computed. The range \( R_j \) for each factor is:
\[
R_j = \max\{k_{j1}, k_{j2}, k_{j3}\} – \min\{k_{j1}, k_{j2}, k_{j3}\}
\]
The results are:
- \( R_A = 264.3 – 241.2 = 23.1 \)
- \( R_B = 258.8 – 247.2 = 11.6 \)
- \( R_C = 257.1 – 250.6 = 6.5 \)
Therefore, the order of influence is: transfer arc current (A) > plasma gas flow (B) > powder gas flow (C). The optimal combination is A₂B₂C₁, i.e., current = 55 A, plasma gas = 1.5 L/min, powder gas = 2.0 L/min. This parameter set yields a hardness closest to the base metal (238.5 HV), and it was used for all subsequent plasma spray welding experiments. This optimization is particularly important when repairing sand foundry defects, because the wrong parameters may lead to excessive dilution and the formation of brittle phases, undermining the repair quality.
4. Microstructural Analysis
4.1 Ni25 Plasma Spray Welding Layer
Figure 1 (not shown here) presents the optical micrographs of the Ni25 plasma spray welding layer at different magnifications. I observed a clear interface between the base metal and the welding layer, indicating sound metallurgical bonding. Since the plasma arc provides high energy density and the composition difference between the Ni25 powder and the cast iron is substantial, elemental diffusion occurs across the interface, forming a strong metallic bond. This is crucial for preventing the recurrence of sand foundry defects at the repaired interface.
The solidification of the molten pool in plasma spray welding is characterized by a relatively low temperature gradient at the liquid-solid interface, resulting in a large constitutional supercooling. This promotes dendritic growth, as seen in the middle and top regions of the welding layer. The primary dendrites form first, followed by secondary arms due to interfacial instability. The microstructural variations across the layer are attributed to differences in cooling rate at different positions.
The X-ray diffraction (XRD) pattern of the Ni25 welding layer (Figure 2, not shown) revealed the presence of the following phases:
\[
\begin{aligned}
&\text{Fe}_{0.64}\text{Ni}_{0.36},\ \text{FeNi}_3,\ \text{Ni},\ \\
&\text{FeCr}_{0.29}\text{Ni}_{0.16}\text{C}_{0.06},\ \text{Ni}_3\text{Si}
\end{aligned}
\]
These phases consist mainly of iron-nickel intermetallics, elemental nickel, and small amounts of complex carbides and nickel silicide. The presence of austenite (stabilized by Ni) is predominant, which contributes to good ductility and toughness, making the layer less susceptible to cracking. This is particularly beneficial when repairing sand foundry defects, because the repair layer must accommodate service stresses without forming new cracks.
Using energy dispersive spectroscopy (EDS) line scanning across the interface, I analyzed the diffusion behavior of elements. The results are illustrated in Figure 3 (not shown). The scan path traversed from the base metal to the welding layer. I found that boron (B) and silicon (Si) showed relatively uniform distributions without significant diffusion. Chromium (Cr), which is absent in the base material, diffused into the base metal side. Nickel (Ni) also diffused into the base metal, while its distribution in the welding layer remained uniform, indicating a high-quality deposit without macrosegregation. Iron (Fe) exhibited a smooth transition across the interface. The diffusion of Ni is particularly important because it promotes graphitization during solidification, thereby suppressing the formation of white iron, which is a common problem when repairing sand foundry defects using conventional arc welding.
4.2 Z116 Electrode Welding Layer
Figure 4 (not shown) presents the microstructures of the Z116 welding layer. The matrix of the weld metal is ferrite, with dispersed precipitates of Fe₃Si, Fe₂SiV, and FeV. During welding, carbon from the base metal diffuses into the weld, while vanadium diffuses toward the fusion line. When carbon and vanadium meet, they form vanadium carbides, which appear as a black band at the fusion boundary. On one hand, this reduces the carbon available for graphitization, promoting the formation of white iron in the semi-melted zone. On the other hand, the combination of C and V helps avoid the formation of high-carbon martensite, partially reducing the hardness of the heat-affected zone. The white iron zone in the semi-melted region was quite wide (approximately 200 μm), consisting of a white matrix of cementite with dark areas of pearlite. This hard and brittle layer is highly undesirable for machining and can cause severe tool damage. The presence of such a wide white iron zone is a direct consequence of using a non-nickel electrode; the high vanadium content does not sufficiently promote graphitization, making it less suitable for repairing sand foundry defects when subsequent machining is required.
The XRD pattern of Z116 layer (Figure 5, not shown) confirmed the phases: Fe, Fe₃Si, Fe₂SiV, FeV. The EDS line scan results (Figure 6, not shown) showed that carbon content fluctuated significantly in the base metal, with high levels corresponding to graphite nodules. In the weld metal, carbon was distributed uniformly. Silicon content in the weld was slightly higher than in the base, which aids in ferrite stabilization. Vanadium was nearly absent in the base metal but present in the weld, confirming its transfer from the electrode. The Fe content was similar in both regions. Manganese content was low and showed little variation, mainly acting as a deoxidizer.
4.3 Z308 Electrode Welding Layer
Figure 7 (not shown) shows the microstructure of the Z308 welding layer. The white matrix is austenite, with black spherical and dot-like graphite particles distributed in the matrix. Additionally, a eutectic of nickel silicide and other compounds was observed at grain boundaries. A narrow white iron zone was present near the fusion line, but its width was significantly smaller than that of the Z116 electrode. The white iron width was around 50 μm and was discontinuous. The smaller width is attributed to the strong graphitizing effect of nickel, which is the main component of Z308. Nickel stabilizes austenite and encourages carbon to precipitate as graphite during cooling, thus reducing the tendency to form cementite (white iron). The XRD pattern (Figure 8, not shown) revealed phases of Fe₀.₆₄Ni₀.₃₆, FeNi₃, C, and Ni₃Si.
EDS line scanning across the Z308 interface (Figure 9, not shown) showed that nickel diffused into the base metal side, which helps to narrow the white iron zone. Carbon content varied across the scan, with high concentrations at graphite locations. The high nickel content in the weld (>30%) ensured that the austenite phase was retained at room temperature, resulting in low hardness and good machinability. Silicon content was relatively uniform, while manganese content was similar in both sides. The presence of graphite in the weld metal is beneficial because its precipitation is accompanied by volume expansion, which reduces residual stresses and helps avoid cracking.
4.4 White Iron Problem in Cast Iron Repair
The formation of white iron in the semi-melted zone is the most critical issue in welding cast iron. White iron is a hard, brittle mixture of cementite and pearlite that forms when carbon precipitates as iron carbide rather than graphite. It is caused by rapid cooling and the absence of sufficient graphitizing elements. In my comparison, I found that plasma spray welding produces no white iron in the semi-melted zone, while both arc welding electrodes produce white iron, with widths of approximately 200 μm for Z116 and 50 μm for Z308.
This difference can be explained by two factors: thermal and metallurgical. The plasma arc has a much higher energy density, which reduces the distance between the liquidus boundary of the molten pool and the solidus line of the gray cast iron. This narrows the semi-melted zone and shortens the time available for carbide formation. Additionally, the Ni25 powder contains nickel, which is a strong graphitizer; the diffusion of nickel into the semi-melted zone further suppresses white iron formation. In contrast, arc welding has a lower energy density, resulting in a wider semi-melted zone and faster cooling, promoting white iron formation. The Z308 electrode, with its high nickel content, produces a narrower white iron zone than Z116, because nickel promotes graphitization more effectively. However, even Z308 cannot eliminate white iron entirely under arc welding conditions.
The presence of white iron increases the hardness of the repaired region to 60 HRC or higher, which complicates subsequent machining. When cutting tools encounter such hard zones, they may experience severe impact or “chatter”, leading to premature tool failure or poor surface finish. Therefore, the elimination of white iron is a major advantage of plasma spray welding for repairing sand foundry defects.
5. Mechanical Properties of the Repaired Layers
5.1 Microhardness Distribution
I measured the microhardness across the cross-section of each repaired specimen, starting from the base metal, through the heat-affected zone, and into the repair layer. The average hardness of the HT250 base metal was 238.5 HV.
For the Ni25 plasma spray welding layer (Figure 10, not shown), the hardness of the repair layer ranged from 230 to 240 HV, which is very close to the base metal hardness. The heat-affected zone hardness was approximately 290 HV, indicating no significant hardening. This is a direct consequence of the absence of white iron.
For the Z308 electrode layer (Figure 11, not shown), the hardness of the weld metal was between 170 and 190 HV, which is lower than the base metal. The semi-melted zone showed a hardness peak of about 456.6 HV, but because the white iron zone was narrow (50 μm), this peak does not seriously affect machinability. The overall hardness profile is acceptable for repair applications.
For the Z116 electrode layer (Figure 12, not shown), the hardness of the weld metal was between 270 and 280 HV, higher than the base metal. The semi-melted zone reached a maximum hardness of 720.9 HV, with a wide white iron zone (200 μm). Such high hardness and wide brittle zone can cause severe tool wear and “chatter” during machining, making Z116 unsuitable for repairing parts that require post-weld finishing.
These results clearly demonstrate that plasma spray welding with Ni25 provides the most favorable hardness profile, closely matching the base cast iron and avoiding the harmful effects of white iron. This is especially important when repairing components subjected to fatigue or abrasive wear, since a hardness mismatch can lead to premature failure or stress concentration at the interface.
5.2 Tensile Bond Strength
I prepared standard tensile test specimens according to GB/T 9439-88. Before welding, a V-shaped groove was machined at the center of each specimen. The groove was then filled using plasma spray welding with Ni25 powder. After welding, the surface was ground to achieve a smooth finish and ensure dimensional accuracy. Tensile testing was performed at room temperature, and the results are summarized in Table 6.
| Specimen No. | Tensile Strength (MPa) | Fracture Location |
|---|---|---|
| 1 | 212.67 | Weld center |
| 2 | 183.34 | Weld center |
| 3 | 196.53 | Weld center |
| 4 | 200.39 | Weld center |
The average tensile strength was calculated as:
\[
\sigma_{avg} = \frac{212.67 + 183.34 + 196.53 + 200.39}{4} = 198.23\ \text{MPa}
\]
All specimens fractured in the weld center, which indicates that the bond strength between the repair layer and the base metal is higher than the tensile strength of the weld metal itself. Therefore, the interface is not the weak link. This confirms that the plasma spray welding layer forms a strong metallurgical bond, which is essential for the long-term reliability of repaired components. In contrast, when sand foundry defects are repaired using methods that produce weak mechanical bonding (e.g., thermal spraying), premature interfacial failure is a common issue. My results show that plasma spray welding can restore the structural integrity of cast iron parts.
5.3 Crack Resistance Test
I evaluated the crack resistance of the repaired layers using the rigid circular disk test. This test imposes a high degree of constraint, which promotes crack formation if the weld metal has poor toughness. I performed continuous bead deposition along the diameter of the disk without preheating or post-weld hammering. Figure 13 (not shown) displays the weld appearance for (a) Ni25 plasma spray welding, (b) Z308 electrode, and (c) Z116 electrode.
For the Ni25 plasma spray welded specimen, the weld appearance was smooth and defect-free, with no cracks detected on the surface or in the cross-section. Similarly, the Z308 electrode specimen showed a good weld appearance without any cracks. However, the Z116 electrode specimen exhibited poor weld appearance, with two visible transverse cracks on the surface. This indicates that the Z116 electrode has a higher tendency for cracking due to the formation of brittle microstructures, such as high hardness white iron and vanadium carbides, which reduce the ductility of the weld metal.
The superior crack resistance of Ni25 plasma spray welding can be attributed to the fine dendritic microstructure of austenite, which is inherently tough, and the absence of white iron at the interface. Additionally, the lower heat input of plasma spray welding results in lower residual stresses. For repairing sand foundry defects, crack resistance is a critical factor because the repaired area may be subjected to cyclic loading or thermal stresses in service.
6. Discussion
The experimental results demonstrate that plasma spray welding with Ni25 powder is the most effective technique among the three methods tested for repairing sand foundry defects in gray cast iron. The main advantages are summarized below:
- No white iron formation: The semi-melted zone remains free of cementite, maintaining machinability.
- Close hardness match: The repair layer hardness is nearly identical to the base metal, avoiding stress concentrations and tool wear.
- High bond strength: The metallurgical bonding provides tensile strengths exceeding 198 MPa.
- Excellent crack resistance: The austenitic microstructure and low residual stress prevent cracking under high restraint.
- Process efficiency: The automated plasma spray welding system offers high deposition rates and consistent quality compared to manual arc welding.
In contrast, Z308 electrode provides acceptable repair quality with a narrow white iron zone, but the lower hardness of the weld metal (170–190 HV) may lead to premature wear in abrasive environments. Z116 electrode is the least suitable due to the wide white iron zone and poor crack resistance. Therefore, for high-quality repair of sand foundry defects, plasma spray welding is the recommended choice, especially when the repaired surface requires post-machining or is exposed to significant mechanical stresses.
It should be noted that the selection of repair technology also depends on the size and location of the defect. For large defects, plasma spray welding may be less economical than manual arc welding, but its superior quality often justifies the cost for critical components. Additionally, the nickel-based alloy powder is more expensive than cast iron electrodes, but the reduced scrap rate and extended service life offset the initial material cost.
Future work could focus on the application of plasma spray welding for in-situ repairs of larger castings and the development of alternative powder compositions that are more cost-effective. Furthermore, the effect of post-weld heat treatment on residual stress relief and microstructural stability could be investigated to further optimize the repair process.
7. Conclusions
Based on my comprehensive study, the following conclusions can be drawn:
- The optimized plasma spray welding parameters for Ni25 powder on HT250 are: transferred arc current of 55 A, plasma gas flow of 1.5 L/min, powder gas flow of 2.0 L/min, voltage of 25 V, shielding gas flow of 12 L/min, and nozzle-to-workpiece distance of 10 mm.
- The Ni25 plasma spray welding layer exhibits a dendritic structure primarily composed of austenite (stabilized by nickel) with minor phases of Fe₀.₆₄Ni₀.₃₆, FeNi₃, Ni, FeCr₀.₂₉Ni₀.₁₆C₀.₀₆, and Ni₃Si. EDS analysis confirmed metallurgical bonding with elemental diffusion across the interface.
- The Z116 welding layer consists of ferrite matrix with dispersed Fe₃Si, Fe₂SiV, and FeV compounds, while the Z308 welding layer is austenitic with graphite precipitates and Ni₃Si eutectic.
- Plasma spray welding completely eliminates white iron in the semi-melted zone, whereas Z308 and Z116 produce white iron zones of about 50 μm and 200 μm, respectively. This is the key factor determining the overall quality of the repair.
- The microhardness of the Ni25 repair layer (235.6 HV) is closest to that of HT250 (238.5 HV), making it the most suitable for applications requiring post-repair machining. The Z308 layer has lower hardness, while the Z116 layer exhibits excessive hardness in the semi-melted zone.
- The average tensile strength of the plasma spray welded specimens is 198.23 MPa, with all fractures occurring in the weld metal, confirming a strong interface bond.
- Crack resistance tests show that Ni25 plasma spray welding and Z308 welding produce crack-free repairs, while Z116 welding leads to visible transverse cracks due to its brittleness.
In summary, plasma spray welding with Ni25 powder is an excellent technology for repairing sand foundry defects in gray cast iron components. It provides a defect-free, machinable, and mechanically sound repair layer that outperforms conventional arc welding with Z116 and Z308 electrodes. The findings of this study offer practical guidance for manufacturing engineers and contribute to the advancement of sustainable repair methods, reducing waste and extending the service life of cast iron parts.