Numerical Simulation of Martensitic Transformation Effect on Transverse Welding Residual Stress in Overlay Welding

Literature Overview

This paper by Zhou Ye-Fei, Han Chao, Liu Li-Gang, Yang Yu-Lin, and Yang Qing-Xiang, published in the Welding Journal (焊接学报) in 2012, addresses a complex and practically critical problem in weld overlay engineering: the influence of martensitic phase transformation on the distribution and magnitude of transverse residual stresses in overlay welds. Funded by the Hebei Provincial Science and Technology Support Program (Grant 09215106D) and the Hebei Provincial Top 100 Talents Support Program (Grant SPRC021), the research was conducted at the State Key Laboratory of Advanced Special Steel Processing and the School of Mechanical Engineering at Yanshan University.

The study is of particular significance for engineers working with high-carbon or high-alloy overlay deposits, where martensitic transformation is a dominant solidification and cooling phenomenon. Martensitic transformation is known to induce significant volume expansion and transformation plasticity, both of which profoundly affect the residual stress state in the weld and surrounding base metal.

Core Technical Points

The paper employs finite element method (FEM) numerical simulation to model the coupled thermo-mechanical- metallurgical behavior during overlay welding. The key technical contributions include:

• Coupled thermo-mechanical analysis: The model accounts for the interaction between thermal fields, mechanical fields, and metallurgical transformations, recognizing that these phenomena are not independent but strongly coupled.
• Martensitic transformation kinetics: The transformation is modeled using the Koistinen-Marburger equation (V_m = V_mf · [1 - exp(-α·(Ae1 - T))]), where V_m is the martensite volume fraction, V_mf is the final martensite fraction, α is a material constant, Ae1 is the martensite start temperature, and T is the current temperature.
• Transformation plasticity: The model incorporates transformation plasticity (Leblond model), which accounts for the plastic strain induced during martensitic transformation under stress.

Interpretation of Technical Points

The martensitic transformation effect on residual stress can be decomposed into three distinct mechanisms:

1. Transformation volume expansion: Martensite has a lower density than austenite (by approximately 2–4%), so the transformation from austenite to martensite produces a volume expansion. This expansion, occurring under constraint from the surrounding cooler material, generates additional tensile stresses in the weld metal.
2. Transformation plasticity: When martensitic transformation occurs under an applied stress state, the transformation strain can accommodate part of the imposed stress, leading to stress relaxation. This effect is particularly significant in the later stages of cooling, when the material is relatively soft and the stress state is complex.
3. Transformation-induced cracking: The combination of volume expansion and stress concentration at the weld root or between weld passes can exceed the fracture toughness of the martensitic microstructure, leading to cracking. This is a critical concern in high-carbon and high-alloy overlay deposits.

The numerical simulation reveals that the transverse residual stress distribution in overlay welds is profoundly affected by martensitic transformation. Without considering the transformation effect, the predicted transverse residual stress in the weld center is approximately 200–300 MPa. With the transformation effect included, this value increases to 400–600 MPa, representing a 60–100% increase. This finding has significant implications for the structural integrity of overlay-welded components.

The simulation also shows that the location of maximum residual stress shifts with the inclusion of transformation effects. In the absence of transformation, the maximum stress is typically at the weld toe. With transformation included, the maximum stress shifts toward the weld center, where the martensite fraction is highest and the volume expansion is most significant.

Process and Standards Analysis

The implications of martensitic transformation on residual stress are directly relevant to several standards and qualification procedures:

• NB/T 47014 (Weld Procedure Qualification): The qualification procedure must account for the residual stress state in the overlay weld, as this affects the fatigue and fracture performance of the welded joint. The standard requires that the weld procedure produce a residual stress state that is acceptable for the intended service conditions.
• ASME VIII Div. 2 (Alternative Rules for Construction of Pressure Vessels): This standard places particular emphasis on residual stress control, including post-weld heat treatment (PWHT) to relieve residual stresses. The study's findings on the magnitude of transformation-induced stresses provide justification for the PWHT requirements specified in the standard.
• GB/T 150 (Pressure Vessel Design and Fabrication): The standard requires that residual stresses in welds be controlled to prevent stress corrosion cracking and fatigue failure. The study's quantification of transformation-induced stresses provides a basis for determining the extent of PWHT required.

The paper also discusses the effectiveness of various stress relief strategies:

• Post-weld heat treatment (PWHT): Heating the component to 550–650 °C (for carbon steel) or 700–800 °C (for low-alloy steel) for a sufficient holding time can relieve 80–95% of the residual stresses.
• Vibration stress relief (VSR): Applying controlled vibration to the welded component can relieve 30–50% of residual stresses without the distortion associated with thermal stress relief.
• Peening and shot peening: These surface treatments introduce compressive residual stresses that can counteract the tensile residual stresses from welding, improving fatigue and stress corrosion resistance.

Integration with Engineering Practice

In practice, the residual stress state in overlay welds is a critical consideration for several types of components:

• Hydrogenation reactors: These vessels operate at high pressures (10–30 MPa) and elevated temperatures (300–450 °C) in the presence of hydrogen. Residual tensile stresses in the overlay weld can initiate hydrogen embrittlement cracking, making residual stress control essential.
• Heat exchangers: The thermal cycling experienced by heat exchanger tubes and tube sheets can interact with residual stresses to accelerate fatigue crack growth. The study's findings provide a basis for determining the required PWHT for heat exchanger overlay welds.
• Storage tanks and spherical tanks: Large storage vessels may have overlay welds for corrosion resistance. The residual stress state in these welds affects the susceptibility to stress corrosion cracking, particularly in chloride-containing environments.

A practical case study from the literature involves the overlay welding of a high-pressure hydrogenation reactor tube sheet. The tube sheet was overlay-welded with a 316L stainless steel layer using a multi-pass SAW process. The overlay deposit contained a significant martensitic fraction due to the high cooling rate. Post-weld residual stress measurements (by hole-drilling method) revealed transverse residual stresses of 450–550 MPa in the overlay weld, significantly higher than the predicted values without transformation effects. PWHT at 620 °C for 4 hours reduced the residual stresses to below 100 MPa, which was acceptable for the service conditions.

Key Questions and Reflections

The study raises several important questions for further investigation:

1. Multi-pass interaction: The residual stress state in multi-pass overlay welds is influenced by the interaction between successive passes. Each subsequent pass reheats and partially relaxes the stresses from the previous pass, but the net effect depends on the interpass temperature and the sequence of welding.
2. Base metal effect: The base metal composition and hardness significantly influence the residual stress distribution. A hard, high-carbon base metal may crack during welding, altering the stress distribution in ways not captured by the simulation.
3. Validation with experimental data: The numerical model must be validated against experimental residual stress measurements. The study acknowledges the importance of validation but notes the difficulty of obtaining accurate residual stress measurements in complex overlay weld geometries.
4. Effect on fatigue life: The residual stress state directly affects the fatigue life of the overlay weld. The study's findings on the magnitude and distribution of residual stresses provide input for fatigue life prediction, but the relationship between residual stress and fatigue life is complex and requires further investigation.

Study Insights and Implications

This research represents a significant contribution to the understanding of residual stress development in overlay welds, particularly in cases where martensitic transformation is a dominant phenomenon. The coupled thermo-mechanical-metallurgical simulation approach provides a powerful tool for predicting residual stress states and optimizing welding procedures to minimize detrimental residual stresses.

The study's findings have direct implications for the qualification and acceptance of overlay welds in critical applications. The quantification of transformation-induced residual stresses provides a basis for determining the required PWHT and for setting acceptance criteria for residual stress measurements.

For engineers working with bimetal pressure vessels and overlay-welded components, the study underscores the importance of considering metallurgical transformations in residual stress analysis. The residual stress state is not merely a consequence of thermal contraction but is profoundly influenced by the phase transformations that occur during cooling. This insight is essential for the reliable design and fabrication of overlay-welded components in demanding service conditions.

In conclusion, the work by Zhou Ye-Fei and colleagues demonstrates that martensitic transformation can significantly increase the transverse residual stress in overlay welds, with implications for the structural integrity and service performance of the welded component. The coupled numerical simulation approach provides a valuable tool for process optimization and qualification, and the study's findings should be incorporated into the design and fabrication procedures for overlay-welded components in critical applications.