In recent years, with the rapid development of the shipbuilding industry, ultra-large container ships exceeding 10,000 TEU have become increasingly popular. The larger scale of vessels necessitates correspondingly larger critical components, such as the rudder horn, which is a vital structural element in ships. The rudder horn, typically made from high-strength steel casting, supports and suspends the rudder system, enduring significant bending fatigue stresses during operation. As a result, cracks often develop on the surface or near-surface regions, posing serious safety risks. In one instance, during a routine inspection of a container ship in dry dock, cracks were discovered at the weld joints between the rudder horn steel casting and the ship’s outer plate. Initial attempts to remove these cracks using conventional carbon arc gouging proved insufficient, as the gouging depth reached 50 mm without eliminating the cracks, halting the welding process. This prompted a comprehensive on-site assessment to ensure the weld performance of the steel casting met stringent requirements. Through meticulous process summarization and material matching tests, welding procedures were validated via welding process trials. By strictly adhering to these procedures during crack repair, we successfully completed the restoration with high quality and efficiency, ensuring excellent weld performance. This experience underscores the complexities involved in repairing ultra-thick steel casting components and highlights the importance of tailored approaches for such critical applications.
The cracks were symmetrically located on both sides of the stern, specifically at the焊缝 between the rudder horn steel casting and the ship’s outer plate. They extended obliquely upward from the weld into the steel casting body. Crack 1 measured approximately 950 mm in length with a maximum depth of 120 mm, while Crack 2 was about 850 mm long and reached a depth of 180 mm. The steel casting itself had a thickness of 380 mm, and the adjacent ship outer plate was made of 45 mm thick Grade E steel. The configuration of these cracks presented significant challenges due to the immense thickness of the steel casting and the deep penetration of the cracks into the material.
The repair of cracks in such ultra-thick steel casting components involves several inherent difficulties. First, steel castings generally have a high carbon equivalent, leading to increased hardenability and susceptibility to cold cracking. The carbon equivalent can be estimated using formulas such as the International Institute of Welding (IIW) formula:
$$C_{eq} = C + \frac{Mn}{6} + \frac{Cr + Mo + V}{5} + \frac{Ni + Cu}{15}$$
For typical steel casting materials like ZG240-450, the carbon content and alloying elements contribute to a higher $C_{eq}$, which exacerbates cracking risks during welding. Second, steel castings often exhibit coarse grain structures and residual stresses from the casting process, making them prone to cold cracks if improper welding materials or techniques are used. Third, inherent casting defects like shrinkage porosity and gas pores can lead to porosity in the weld metal due to gas decomposition during arc welding. Fourth, the substantial thickness of the steel casting (380 mm) makes rapid heating difficult, complicates stress relief, and increases the likelihood of cracking under restrained conditions. Fifth, the depth of the cracks (up to 180 mm) is unprecedented in repair scenarios, with no prior documented cases for such thick steel casting repairs. Sixth, environmental factors, such as working on water in windy and cold conditions (e.g., April weather), accelerate heat dissipation, making it challenging to maintain interpass temperatures and further increasing crack susceptibility. Therefore, a well-defined and rigorously controlled welding process is essential to prevent defects like cracks in the weld and heat-affected zone (HAZ).
Prior to initiating the repair, welding procedure qualifications were necessary. However, due to time constraints and the lack of existing qualifications for such thick steel castings, we referenced a third-party shipyard’s procedure qualification for a 40 mm thick marine steel casting welded to high-strength steel EH36. We analyzed the chemical compositions of the materials involved—ZG240-450 steel casting and E36 plate—to ensure compatibility. The chemical compositions are summarized in the tables below.
| Element | C | Si | Mn | P | S | Ni | Cr | Mo | Al | Cu |
|---|---|---|---|---|---|---|---|---|---|---|
| Standard | ≤0.23 | ≤0.60 | 0.7~1.6 | ≤0.035 | ≤0.035 | ≤0.4 | ≤0.3 | ≤0.15 | 0.015~0.08 | ≤0.3 |
| Actual | 0.182 | 0.476 | 1.031 | 0.018 | 0.018 | 0.054 | 0.405 | 0.0059 | 0.02 | 0.045 |
| Element | C | Si | Mn | P | S | Al | Cr | Mo | Ni | Cu | Ceq |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Standard | ≤0.18 | 0.1~0.35 | 0.7~1 | ≤0.035 | ≤0.035 | ≥0.02 | ≤0.20 | 0.08 | 0.40 | 0.35 | 0.4 |
| Actual | 0.12 | 0.23 | 1.13 | 0.015 | 0.0021 | 0.031 | 0.20 | 0.035 | 0.40 | 0.35 | 0.3 |
The carbon equivalent for the E36 plate was calculated using the IIW formula, resulting in a value of approximately 0.3%, which is within acceptable limits for welding. Based on this analysis, welding parameters were established to minimize the heat-affected zone and prevent defects. The welding procedure involved shielded metal arc welding (SMAW) for initial layers and flux-cored arc welding (FCAW) for subsequent layers, as detailed in the following table.
| Weld Pass | Process | Filler Material (mm) | Welding Current (A) | Arc Voltage (V) | Polarity | Welding Speed (cm/min) |
|---|---|---|---|---|---|---|
| 1~2 | SMAW | CHE50 (φ3.2) | 100 | 23~25 | DCEP | 5~6 |
| 3~5 | SMAW | CHE50 (φ4.0) | 140~150 | 24~26 | DCEP | 6~9 |
| 6~7 | FCAW | SQJ-501 (φ1.2) | 160~210 | 25~28 | DCEP | 8~9 |
| 8~19 | FCAW | SQJ-501 (φ1.2) | 160~210 | 25~28 | DCEP | 13~21 |
| ≥20 | FCAW | SQJ-501 (φ1.2) | 160~210 | 25~28 | DCEP | 11 |
This procedure was later submitted to the BV classification society for approval on another vessel with similar cracks, and the results fully met the requirements, validating our approach for steel casting repairs.
The crack repair process was divided into two phases: dry dock operations for crack removal and water-based operations for welding. The entire sequence included preheating, carbon arc gouging to eliminate cracks, slow cooling, grinding, non-destructive testing (NDT), preheating again, welding, post-weld heat treatment (PWHT), and final NDT. Each step was critical to ensure the integrity of the steel casting repair.
Preheating was essential due to the high carbon equivalent, thickness, and restraint of the steel casting. Conventional flame heating was inadequate, so we employed ceramic electric heating pads controlled by a WSK intelligent program temperature control unit. This system allowed precise temperature management with feedback from thermocouples. Before gouging, heating pads were placed around the crack areas, as shown in the setup. The temperature was set to 300°C, with a ramp-up time of 0.5–1 hour and a soak time of 1–2 hours to allow heat penetration into the steel casting body. The preheating helps reduce the cooling rate and minimize thermal stresses, which can be expressed by the formula for thermal stress:
$$\sigma = E \alpha \Delta T$$
where $\sigma$ is the thermal stress, $E$ is Young’s modulus, $\alpha$ is the coefficient of thermal expansion, and $\Delta T$ is the temperature difference. By preheating, we reduced $\Delta T$ and thus the risk of cracking.
Carbon arc gouging was used to remove the cracks entirely. Given the depth, a U-shaped groove was preferred to reduce weld metal volume. Flat carbon rods were employed to avoid sharp corners, and the groove ends were tapered at four times the plate thickness. For Crack 2, which was exceptionally deep, a section of the outer plate (approximately 400 mm × 600 mm) was removed to facilitate access. After gouging, the surfaces were ground to remove the carburized layer (high-carbon grains) that could cause defects. The grooves were then cooled slowly to below 50°C at a rate of 50°C/h before conducting penetrant testing (PT) or magnetic particle testing (MT) to confirm complete crack elimination.
Welding was performed on water with environmental protections, including scaffolding and double-layer plastic sheets for wind and rain shielding, along with insulation blankets to maintain temperature. Preheating was re-established to 300°C using ceramic heaters, with a soak time of 1–2 hours. Upon reaching temperature, the heaters were removed from the groove surfaces, and welding commenced simultaneously on both cracks. Due to the groove depth, shielded metal arc welding (SMAW) with low-hydrogen electrodes (J507, equivalent to AWS E7015) was used for the initial layers. Electrodes were baked at 350°C for 2 hours and stored in a 100°C holding oven. The first two passes used φ3.2 mm electrodes to create a smooth groove base, followed by φ4.0 mm or φ5.0 mm electrodes for multilayer, multipass welding. Each pass overlapped the previous by 1/3 to 1/2 width to ensure proper fusion. With approximately 550 passes for the deeper crack, sequencing was crucial to manage residual stresses. The welding direction and layer sequence were alternated as illustrated: for each layer, welding started from the center and proceeded outward, and layers were stacked in a staggered pattern. After each pass, peening with a pneumatic needle scaler was performed to relieve stress. Interpass temperature was monitored with a pyrometer and maintained at or above 150°C to prevent cold cracking. The heat input during welding must be controlled to avoid excessive grain growth; it can be calculated as:
$$Q = \frac{60 \times I \times V}{1000 \times S}$$
where $Q$ is the heat input (kJ/mm), $I$ is the current (A), $V$ is the voltage (V), and $S$ is the welding speed (mm/min). For our parameters, $Q$ was kept within safe limits to preserve the steel casting’s microstructure.
Post-weld heat treatment was conducted immediately after welding to allow hydrogen diffusion and stress relief. The welded area was insulated with ceramic heaters and blankets, heated from 300°C to 600–650°C at 50°C/h, held for 4 hours (extended due to the steel casting thickness), and cooled to 300°C at 50°C/h before natural cooling to room temperature over 24 hours. This process softens the HAZ and reduces hardness, mitigating delayed cracking. The tempering effect can be described by the Larson-Miller parameter for creep resistance, but in this context, the focus was on stress relaxation.
Non-destructive testing was performed 48 hours after welding to detect any defects. For the weld area between the outer plate and steel casting (Region A), ultrasonic testing (UT) with shear wave probes was used. For the repaired steel casting region (Region B), a combination of dual-crystal longitudinal wave probes for near-surface defects and straight beam probes for internal defects was employed, supplemented by angle beam probes for cross-checking. Surface inspection via PT or MT ensured no open defects. The UT inspection followed standard codes, with sensitivity calibrated for the steel casting’s acoustic properties. The results showed no indications of cracks or other discontinuities, confirming the success of the repair.
In conclusion, the repair of cracks in ultra-thick steel casting components like rudder horns demands a systematic approach tailored to the material’s characteristics. Key factors include thorough preheating, controlled welding parameters, stress management through peening and PWHT, and rigorous NDT. This project not only restored the structural integrity of the vessel but also provided valuable insights for future steel casting repairs in maritime applications. The use of advanced heating systems, proper filler material selection, and adherence to validated procedures were instrumental in overcoming the challenges posed by the steel casting’s thickness and environmental constraints. As ships continue to grow in size, such repairs will become more common, necessitating continued innovation in steel casting welding technology.
Throughout this process, the importance of understanding steel casting behavior cannot be overstated. Steel castings, due to their manufacturing process, exhibit unique metallurgical properties that must be accounted for in repair schemes. By leveraging formulas for carbon equivalent and heat input, along with empirical data from procedure qualifications, we can optimize welding practices for steel castings. Future work may involve developing more accurate models for thermal stress distribution in thick steel castings or exploring alternative welding methods like narrow-gap welding to reduce repair time. Ultimately, the lessons learned here contribute to the broader knowledge base on maintaining and repairing critical steel casting components in the marine industry.