In our research, we focused on the welding of a large steel casting component used in offshore equipment, specifically examining the dissimilar joint between NORSOK M-122 Gr420E steel casting and F36 low-alloy steel. The steel casting, with a thickness of 150 mm, and the F36 plate, at 75 mm, presented significant challenges due to their differences in chemical composition, mechanical properties, and thermal characteristics. Steel castings are commonly employed in high-load, complex structural nodes, but their inherent issues, such as high carbon equivalent, coarse microstructure, and poor density, make welding more difficult compared to rolled carbon steels. As offshore platforms evolve toward larger and more complex designs, the application of steel castings has become increasingly prevalent, necessitating advanced welding techniques to ensure joint integrity. This study details our comprehensive approach to developing a reliable welding procedure, including material analysis, process optimization, and rigorous testing to meet the stringent requirements of modern engineering projects.
The primary difficulty in welding steel castings lies in their high carbon content and thick sections, which can lead to cracking and reduced toughness in the heat-affected zone (HAZ). For instance, the carbon equivalent (Ceq) is a critical parameter in assessing weldability. Generally, when Ceq ≤ 0.4%, the material exhibits good weldability; for 0.4% < Ceq < 0.6%, weldability is moderate; and when Ceq > 0.6%, welding becomes highly challenging. In our case, the steel casting had a Ceq of 0.479%, indicating moderate weldability that required careful control of thermal conditions. We employed the following formula to calculate carbon equivalent, which is widely used in welding metallurgy:
$$C_{eq} = C + \frac{Mn}{6} + \frac{Cr + Mo + V}{5} + \frac{Ni + Cu}{15}$$
Where the elemental concentrations are in weight percent. This formula helped us evaluate the susceptibility to cold cracking and guided our preheating and interpass temperature strategies. Additionally, the dissimilar nature of the joint, combining steel casting with low-alloy steel, introduced complexities due to differing coefficients of thermal expansion and mechanical properties, potentially leading to residual stresses and distortion. Our objective was to optimize the welding process to achieve a joint that meets the demanding standards for strength, toughness, and crack resistance, as specified by project requirements, including Charpy V-notch impact energy and crack tip opening displacement (CTOD) values.
To begin, we conducted a detailed analysis of the base materials. The steel casting, conforming to NORSOK M-122 Gr420E, was supplied in a quenched and tempered condition. This steel casting is typically used in critical offshore structures due to its high strength and toughness. The component we examined was a tubular support structure approximately 3.6 m in diameter, with a maximum wall thickness of 450 mm and a weight of around 130 tons. For our welding procedure qualification, we used a test plate of the same steel casting material, with thicknesses ranging from 150 mm to 75 mm to simulate the actual production geometry. The chemical composition and mechanical properties of this steel casting were meticulously evaluated, as summarized in the tables below.
| Element | C | S | P | Mn | Si | Al | Sn | As |
|---|---|---|---|---|---|---|---|---|
| Standard Value | ≤0.140 | ≤0.008 | ≤0.012 | ≤1.600 | ≤0.600 | ≤0.050 | ≤0.015 | ≤0.020 |
| Measured Value | 0.097 | 0.001 | 0.008 | 1.402 | 0.497 | 0.041 | 0.008 | 0.006 |
| Element | Ni | Mo | Cu | V | Ti | Sb | Cr | – |
| Standard Value | ≤1.500 | ≤0.300 | ≤0.250 | ≤0.050 | ≤0.0400 | ≤0.0100 | ≤0.250 | – |
| Measured Value | 1.363 | 0.085 | 0.134 | 0.024 | 0.0017 | 0.0011 | 0.135 | – |
The mechanical properties of the steel casting were tested at different thickness locations to account for potential inhomogeneities inherent in casting processes. For example, tensile tests were conducted at the 1/2 and 3/4 thickness positions, and hardness measurements were taken across the cross-section. The results confirmed that the steel casting met the required specifications, with yield strength above 420 MPa, tensile strength over 540 MPa, and impact energy at -40°C exceeding 42 J. The CTOD value, a key indicator of fracture toughness, was measured at 0.57 mm, well above the minimum requirement of 0.25 mm. This comprehensive characterization ensured that the steel casting was suitable for the welding process.
| Property | Yield Strength (MPa) | Tensile Strength (MPa) | Impact Energy at -40°C (J) | Elongation (%) | CTOD (mm) |
|---|---|---|---|---|---|
| Standard Value | ≥420 | ≥540 | ≥42 | ≥20 | ≥0.25 |
| Measured Value | 468 | 583 | 106 | 26 | 0.57 |
Next, we examined the F36 low-alloy steel, which was supplied in a normalized condition according to GB/T 712-2022. This material is known for its good toughness and weldability, making it a common choice in marine structures. The chemical composition and mechanical properties of F36 are presented in the following tables. The carbon equivalent for F36 was lower than that of the steel casting, reducing the risk of welding-related issues. However, the dissimilar joint required careful consideration of thermal expansion mismatches and potential stress concentrations.
| Element | C | S | P | Mn | Si | Cr | Mo | Ni | Cu | Al |
|---|---|---|---|---|---|---|---|---|---|---|
| Standard Value | ≤0.160 | ≤0.025 | ≤0.025 | ≤1.600 | ≤0.500 | ≤0.200 | ≤0.080 | ≤0.800 | ≤0.350 | ≥0.020 |
| Measured Value | 0.120 | 0.0015 | 0.011 | 1.560 | 0.300 | 0.040 | 0.010 | 0.180 | 0.030 | 0.040 |
| Property | Yield Strength (MPa) | Tensile Strength (MPa) | Impact Energy at -60°C (J) | Elongation (%) |
|---|---|---|---|---|
| Standard Value | ≥355 | 490-620 | ≥50 | ≥21 |
| Measured Value | 416 | 542 | 222 | 31.5 |
Moving to the welding process, we selected flux-cored arc welding (FCAW) as the primary method due to its efficiency and suitability for thick sections. This process allows for high deposition rates and deep penetration, which are advantageous for large steel casting components. We used an 80% Ar + 20% CO2 shielding gas mixture to minimize oxidation and enhance arc stability, thereby improving the toughness of the weld metal. The welding consumable chosen was DW-A55LSR (AWS A5.29 E81T1-Ni1M), which provides excellent low-temperature toughness and strength matching the base materials. The nickel content in the filler metal helps to counteract the brittleness often associated with steel castings.
For the joint design, we opted for a double-sided 40° groove with a root gap of 3-5 mm and a root face of 1-2 mm. This configuration reduces the volume of weld metal required and minimizes residual stresses. The groove was prepared by mechanical machining to ensure accuracy and avoid defects from thermal cutting. The welding parameters were meticulously controlled, with a focus on heat input, which we kept below 1.8 kJ/cm per layer to prevent excessive grain growth and HAZ degradation. The heat input (HI) can be calculated using the formula:
$$HI = \frac{60 \times V \times I}{1000 \times S}$$
Where V is the voltage in volts, I is the current in amperes, and S is the travel speed in mm/min. For example, in the horizontal position, we maintained currents of 220-250 A for the root pass, 230-270 A for fill passes, and 230-250 A for the cap pass, with corresponding voltages of 24-26 V, 25-28 V, and 25-27 V, respectively. Travel speeds were adjusted between 250-450 mm/min depending on the pass and position. Similarly, for vertical-up welding, lower currents and voltages were used to control fluidity and prevent sagging.
Temperature control was critical throughout the process. We preheated the joint to at least 150°C using resistance heating blankets, ensuring uniformity within a 75 mm zone on either side of the groove. The interpass temperature was strictly maintained between 150°C and 220°C to avoid rapid cooling and hydrogen-induced cracking. Post-weld, the assembly was insulated with heat-resistant blankets to allow slow cooling to room temperature, further reducing thermal stresses. We also implemented strict procedures for interpass cleaning, avoiding pneumatic tools to prevent mechanical shock that could initiate cracks in the steel casting HAZ.
During welding, we emphasized technique consistency, such as using a weaving motion with pauses at the groove edges to ensure sidewall fusion. Each layer was limited to a thickness of about 5 mm and a width of no more than 16 mm to manage dilution and microstructural homogeneity. We conducted trial runs on mock-ups to calibrate parameters before proceeding to the actual test plates, recording all data for reproducibility. Gas flow rates were maintained at 15-25 L/min, and cylinders were replaced when pressure dropped below 5 MPa to avoid contamination.
After welding, the test plates underwent non-destructive testing, including ultrasonic and magnetic particle inspections, which confirmed the absence of defects like cracks or lack of fusion. We then performed a series of mechanical tests to evaluate the joint performance. Transverse tensile tests showed strengths of 561 MPa for horizontal welds and 559 MPa for vertical-up welds, with failures occurring in the F36 base metal, indicating overmatching weld strength. All-weld-metal tensile tests yielded strengths of 626 MPa and 617 MPa for the respective positions, satisfying the requirements. Bend tests, conducted with a 40 mm diameter plunger and 180° angle, revealed no surface defects, demonstrating good ductility.
Charpy V-notch impact tests were carried out at -40°C for the weld and steel casting HAZ, and at -60°C for the F36 base metal. Specimens were taken from near the surface and mid-thickness locations, including the weld center, fusion line, and HAZ regions. The average impact energies exceeded 90 J in all cases, far above the minimum of 42 J, highlighting the excellent toughness achieved. Macrostructural examination of cross-sections showed sound weld beads without imperfections, and hardness surveys indicated values within acceptable limits, as detailed below.
| Location | Steel Casting Base Metal | Steel Casting HAZ | Weld Center | F36 HAZ | F36 Base Metal |
|---|---|---|---|---|---|
| Near Upper Surface | 190 | 272 | 210 | 261 | 183 |
| Weld Root | 206 | 258 | 230 | 228 | 185 |
| Near Lower Surface | 204 | 252 | 218 | 242 | 179 |
Finally, we conducted CTOD tests on full-thickness specimens from the vertical-up weld, which had the highest heat input. Three samples were prepared with fatigue-pre cracked notches in the weld center and tested at -10°C. The results, with CTOD values of 0.740 mm, 0.872 mm, and 0.754 mm, demonstrated superior resistance to crack propagation, well above the 0.25 mm threshold. This confirms that our welding process effectively mitigates the risk of brittle fracture in critical applications involving steel castings.
In conclusion, our study successfully developed a robust welding procedure for joining heavy steel casting components to low-alloy steel plates. By optimizing parameters such as heat input, preheating, and interpass temperature, we achieved joints with high strength, toughness, and crack resistance. This approach has been validated in production, where a 135-meter-long weld achieved a first-pass acceptance rate of 96.7%, with any defects remedied in a single repair. The findings provide a reliable framework for similar applications, emphasizing the importance of meticulous process control in welding steel castings. Future work could explore automated welding techniques or alternative filler metals to further enhance efficiency and performance in offshore structures.