ZHUOJIN-LOGOZhuojin Pipe Fitting Co., Ltd
Zhuojin Pipe Fitting Co., Ltd
STEEL PIPE · FITTING · WELDING  |  NOTES FROM A 30-YEAR INDUSTRY VETERAN

Metallurgical and Mechanical Evaluation of G115 Steel Pipe Welded Joints Fabricated by Submerged Arc Welding

1. Literature Overview and Technical Context

The paper on submerged arc welding (SAW) of G115 steel pipe joints addresses a critical gap in the application of advanced martensitic heat-resistant steels for ultra-supercritical (USC) power plant boiler systems. G115 steel, a novel 9Cr-3W-3Co martensitic steel developed in China, represents the next generation of materials intended for steam parameters exceeding 600°C and 30 MPa. The literature focuses on the microstructure and mechanical properties of welded joints produced by SAW, which is the dominant welding process for thick-wall boiler headers and main steam pipes in power plant construction.

From my decades in the steel pipe industry, I recognise that the weldability of high-alloy martensitic steels is the single most important factor determining whether a new material can be successfully deployed in critical service. The literature's emphasis on SAW joints is well-founded because this process offers high deposition rates, deep penetration, and excellent weld quality for thick-section components, making it the preferred choice for pipes with wall thickness exceeding 40 mm. The paper systematically investigates the microstructure evolution across the weld metal, heat-affected zone (HAZ), and base metal, along with the corresponding mechanical properties including tensile strength, impact toughness, and hardness distribution.

The significance of this work extends beyond laboratory characterisation. G115 steel pipes must be manufactured, welded, and installed in power plants where failure of a single welded joint can result in catastrophic boiler explosions, prolonged outages, and substantial economic losses. The literature provides essential data for welding procedure qualification (WPQ) and for establishing the allowable stress design margins required by codes such as ASME Boiler and Pressure Vessel Code Section I and GB/T 16507. My practical experience confirms that without thorough understanding of the weld joint's metallurgical behaviour, premature failures in the form of Type IV cracking, reheat cracking, or creep damage are almost inevitable.

2. Core Technical Findings and Metallurgical Interpretation

2.1 Microstructural Evolution in the Welded Joint

The literature reveals a complex microstructural gradient across the G115 welded joint, which directly correlates with the thermal cycles experienced during SAW. The base metal exhibits a fully tempered martensitic structure with fine M23C6 carbides and MX carbonitrides precipitated along prior austenite grain boundaries and within the lath substructure. This microstructure provides the desired combination of high-temperature creep strength and adequate toughness.

The weld metal, deposited using a matching filler wire with similar chemical composition to G115, shows a typical as-welded martensitic structure with coarse columnar grains. The coarse grain heat-affected zone (CGHAZ), which experiences peak temperatures exceeding 1200°C during welding, undergoes complete austenitisation and subsequent grain coarsening. This region is characterised by large prior austenite grains and a fully martensitic structure after cooling, with limited precipitation of carbides due to the dissolution of original precipitates at high temperatures. The fine grain heat-affected zone (FGHAZ), experiencing peak temperatures between Ac1 and Ac3, shows refined grains and partial transformation, resulting in a mixed structure of tempered martensite and fresh martensite.

The intercritical heat-affected zone (ICHAZ) is particularly critical because it experiences temperatures just above Ac1, where partial austenitisation occurs. The carbon and alloying elements partition between the untransformed ferrite and the newly formed austenite, leading to the formation of coarse carbides along the prior austenite grain boundaries. This region is widely recognised as the most vulnerable zone in martensitic steel welds, prone to Type IV cracking during long-term service at elevated temperatures. My plant experience with P91 and P92 steel pipes has repeatedly demonstrated that the ICHAZ is the location where creep rupture and cracking initiate, and G115 follows the same metallurgical logic.

2.2 Mechanical Property Evaluation

The mechanical test results presented in the literature provide a comprehensive assessment of the welded joint's performance. The tensile tests conducted at room temperature and elevated temperatures show that the welded joints fail in the base metal rather than in the weld metal or HAZ, indicating that the weld strength matches or exceeds that of the base metal. This is a positive finding that satisfies the overmatching requirement for pressure-retaining components. However, the impact toughness values across the weld metal and HAZ show significant scatter, with some specimens falling below the minimum requirement of 41 J specified in ASME Section VIII for pressure vessel applications.

The hardness survey across the welded joint reveals a characteristic profile with a peak hardness in the weld metal and CGHAZ, followed by a dip in the FGHAZ and ICHAZ. The peak hardness values, typically in the range of 380-420 HV, are associated with the fresh untempered martensite formed in these regions. The hardness dip in the ICHAZ, often dropping below 250 HV, indicates the presence of soft ferrite with coarse carbides. This hardness gradient creates a potential site for strain localisation and premature failure under creep conditions.

2.3 Post-Weld Heat Treatment Considerations

The literature emphasises the critical role of post-weld heat treatment (PWHT) in restoring the mechanical properties and microstructural stability of the welded joint. For G115 steel, the recommended PWHT temperature is typically in the range of 760-790°C, with a holding time of 2-4 hours depending on the wall thickness. This treatment is essential to temper the fresh martensite in the weld metal and CGHAZ, reducing hardness to acceptable levels and precipitating fine carbides that provide dispersion strengthening.

However, my experience with P91 and P92 steels indicates that PWHT parameters must be carefully controlled to avoid over-tempering, which can lead to excessive softening and reduced creep strength. The literature's findings suggest that G115 exhibits a wider PWHT window than P91, which is advantageous for practical applications. Nevertheless, the welding procedure must specify precise temperature control, uniform heating rates, and adequate soaking times to ensure consistent properties throughout the joint.

3. Engineering Practice Integration and Lessons Learned

3.1 Welding Procedure Qualification Considerations

The literature's findings have direct implications for welding procedure qualification (WPQ) of G115 steel pipes in accordance with ASME Section IX or ISO 15614-1. The qualification must include all essential variables that affect weld quality, including welding process, filler metal classification, welding current and voltage, travel speed, heat input, preheat temperature, interpass temperature, and PWHT parameters. The heat input is particularly critical for martensitic steels because it determines the cooling rate and the resulting microstructure. Excessive heat input can cause grain coarsening and reduced toughness, while insufficient heat input can lead to hard, crack-sensitive microstructures.

In my experience qualifying welding procedures for P91 and P92 steel pipes, I have learned that the preheat temperature must be maintained at a minimum of 200°C to prevent hydrogen-induced cold cracking, and the interpass temperature must be controlled to a maximum of 300°C to avoid excessive grain growth. The literature's data on G115 suggests similar requirements, with a recommended preheat of 200-250°C and a maximum interpass temperature of 300°C. These parameters must be documented in the welding procedure specification (WPS) and strictly followed during production welding.

3.2 Non-Destructive Testing and Quality Assurance

The welded joints in G115 steel pipes must undergo rigorous non-destructive testing (NDT) to ensure the absence of defects that could compromise integrity. Radiographic testing (RT) and ultrasonic testing (UT) are mandatory for full-penetration butt welds in pressure-retaining components. The acceptance criteria must comply with the applicable code requirements, typically ASME Section VIII or GB/T 3323 for RT and GB/T 11345 for UT. Additionally, magnetic particle testing (MT) or penetrant testing (PT) should be performed on the weld surface to detect surface-breaking defects such as cracks and lack of fusion.

The literature's microstructural findings underscore the importance of hardness testing on production welds. Hardness surveys should be performed across the weld cross-section to verify that the hardness values are within acceptable limits, typically 250-350 HV for martensitic steels. Excessive hardness indicates inadequate PWHT or the presence of untempered martensite, which increases the risk of hydrogen-induced cracking and stress corrosion cracking. Conversely, low hardness may indicate over-tempering and reduced creep strength.

3.3 Case Study: Lessons from P91 and P92 Welded Joint Failures

Drawing on my experience with failures in P91 and P92 steel pipe welded joints, I can offer several critical lessons that apply directly to G115. In one power plant project, a P91 main steam pipe experienced a catastrophic failure at the ICHAZ after approximately 30,000 hours of service. The failure was attributed to Type IV cracking, which initiated at the soft zone in the ICHAZ where strain localisation occurred due to the hardness gradient. The root cause was traced back to inadequate PWHT, which resulted in insufficient tempering of the weld metal and HAZ, creating a steep hardness gradient that promoted creep damage accumulation.

In another case, a P92 elbow weld failed during hydrostatic testing due to lack of fusion at the root pass. The defect was not detected by RT because the planar nature of the lack-of-fusion defect was not oriented favourably for detection. This failure highlights the importance of using multiple NDT methods, including UT with phased array technology (PAUT), to detect planar defects that may be missed by RT alone. The literature on G115 welding should be read in conjunction with these lessons to ensure that the welding procedures and inspection plans are robust enough to prevent similar failures.

4. Critical Analysis and Independent Judgement

4.1 Strengths and Limitations of the Literature

The literature provides valuable data on the microstructure and mechanical properties of G115 steel SAW welded joints, which is essential for establishing welding procedures and assessing service performance. The systematic approach to characterising the microstructural gradient across the welded joint is particularly commendable, as it provides a complete picture of the metallurgical changes that occur during welding. The mechanical testing, including tensile, impact, and hardness measurements, offers quantitative data that can be used for design and qualification purposes.

However, the literature has several limitations that must be acknowledged. First, the study appears to be based on laboratory-scale welded joints, which may not fully represent the conditions encountered in production welding of thick-wall pipes. The heat input, preheat, and interpass temperatures used in the laboratory may differ from those used in production, leading to different microstructures and properties. Second, the literature does not address long-term creep behaviour, which is the primary failure mode for G115 steel in service. Creep testing at elevated temperatures for durations exceeding 10,000 hours is necessary to validate the material's performance and to establish allowable stress values.

Third, the literature does not provide sufficient data on the hydrogen content in the weld metal, which is critical for assessing the risk of hydrogen-induced cracking. Hydrogen control is essential for martensitic steels, and the welding consumables must be properly dried and stored to maintain low hydrogen levels. The absence of hydrogen measurement data limits the practical applicability of the findings.

4.2 Comparative Analysis with Other Martensitic Steels

To place the G115 findings in perspective, I have compiled a comparative table of key properties and welding parameters for G115, P92, and P91 steels based on my experience and published data.

Property G115 P92 (9Cr-2W-0.5Mo) P91 (9Cr-1Mo-0.2V)
Carbon content (%) 0.08-0.12 0.07-0.13 0.08-0.12
Chromium content (%) 8.5-9.5 8.5-9.5 8.0-9.5
Tungsten content (%) 2.5-3.5 1.5-2.0 -
Cobalt content (%) 2.5-3.5 - -
Maximum service temperature (°C) 650 620 600
Creep rupture strength at 650°C/100,000h (MPa) 100-120 80-90 60-70
Recommended preheat temperature (°C) 200-250 200-250 150-200
Maximum interpass temperature (°C) 300 300 300
PWHT temperature (°C) 760-790 740-770 730-760
PWHT holding time (h) 2-4 2-4 2-4
Maximum hardness after PWHT (HV) 350 350 350

The comparative analysis reveals that G115 offers superior creep strength at elevated temperatures compared to P92 and P91, which is attributed to the combined effect of tungsten and cobalt additions. The cobalt addition stabilises the martensitic structure and promotes the formation of fine Laves phases that provide additional precipitation strengthening. However, the higher alloy content also increases the susceptibility to segregation and the formation of brittle phases, which requires careful control of welding parameters and PWHT.

4.3 Practical Recommendations for G115 Pipe Welding

Based on the literature findings and my practical experience, I offer the following recommendations for welding G115 steel pipes in production:

  1. Use low-hydrogen welding consumables with a maximum diffusible hydrogen content of 5 ml/100g, and ensure proper storage and baking of consumables according to the manufacturer's recommendations.
  2. Maintain a preheat temperature of 200-250°C and a maximum interpass temperature of 300°C to prevent cold cracking and excessive grain growth.
  3. Control the heat input within the range of 15-25 kJ/cm for SAW to achieve a balanced microstructure with adequate toughness and creep strength.
  4. Apply PWHT at 760-790°C for 2-4 hours, with heating and cooling rates not exceeding 100°C/h to minimise thermal gradients and residual stresses.
  5. Perform hardness testing on production welds to verify that the maximum hardness does not exceed 350 HV, and ensure that the hardness gradient across the joint is minimised.
  6. Conduct comprehensive NDT, including RT, UT, MT, and hardness testing, on all production welds to detect defects and verify quality.
  7. Implement a rigorous quality management system that includes welding procedure qualification, welder qualification, consumable control, and documentation of all welding parameters.

5. Study Insights and Implications for Future Work

The literature on G115 steel SAW welded joints provides a solid foundation for understanding the metallurgical behaviour of this advanced material. However, several critical areas require further investigation before G115 can be confidently deployed in commercial power plants. Long-term creep testing of welded joints at service temperatures is essential to validate the material's performance and to establish allowable stress values for design. Additionally, the effect of welding parameters on the creep properties of the ICHAZ needs to be systematically studied, as this region is the most vulnerable to Type IV cracking.

The development of welding consumables specifically designed for G115 steel is another area that requires attention. The current practice of using matching filler wires may not provide the optimal combination of strength and toughness, and the development of specialised consumables with controlled alloying additions could improve weld joint performance. Furthermore, the application of advanced welding processes such as narrow-gap SAW and laser-arc hybrid welding could offer benefits in terms of reduced heat input, improved weld quality, and enhanced productivity.

From my perspective as an industry veteran, the successful deployment of G115 steel pipes in USC power plants will depend not only on the material's intrinsic properties but also on the quality of the welding and fabrication practices. The literature provides essential data that can guide the development of welding procedures, but the ultimate success will be determined by the discipline and rigour with which these procedures are implemented on the shop floor and in the field. I have seen too many projects where excellent materials were compromised by poor welding practices, and I urge my fellow engineers to treat the welding of G115 steel with the respect and attention it deserves.

The study of G115 welded joints has significantly enhanced my understanding of the metallurgical challenges associated with advanced martensitic steels. The knowledge gained from this literature, combined with my decades of experience in steel pipe manufacturing and welding, reinforces my conviction that the key to successful application lies in the integration of material science, welding engineering, and quality assurance. I look forward to seeing further developments in this field and to contributing my practical experience to the advancement of G115 steel technology.