Comparative Numerical Simulation of Laser and TIG Welding of IN738 Superalloy Thin Plate: Residual Stress and Plastic Strain Analysis
Literature Overview and Core Findings
This paper from Xi'an Thermal Power Research Institute presents a finite element analysis of laser welding and TIG welding of IN738 high-temperature superalloy thin plates using MSC.Marc software. The authors calculated temperature fields and then derived residual stress and plastic strain distributions for both processes, validating the simulations against experimental measurements. The key findings are that laser welding produces higher peak residual stress than TIG welding, with higher longitudinal residual stress in the weld centerline and heat-affected zone. However, the tensile stress zone in laser welds is narrower, and the plastic deformation zone is smaller with lower plastic strain values. The calculated temperature fields and residual stress fields showed good agreement with experimental measurements, except for the transverse residual stress in the weld itself, where significant differences were observed.
Technical Interpretation and Process Implications
IN738 is a precipitation-hardened nickel-based superalloy used extensively in gas turbine components, and it is notoriously difficult to weld due to its high susceptibility to hot cracking and strain age cracking. The paper's comparison of laser and TIG welding is directly relevant to the repair and fabrication of turbine components, where both processes are used. The finding that laser welding produces higher peak residual stress but a narrower stress zone is a classic trade-off. Laser welding concentrates heat input in a small area, creating steep temperature gradients and high localised stresses, but the overall heat-affected zone is smaller, which reduces the extent of the degraded region. TIG welding, with its broader heat input, distributes the stress over a wider area but creates a larger heat-affected zone with more extensive microstructural changes. For IN738, which is sensitive to strain age cracking in the heat-affected zone, the choice between laser and TIG is not straightforward. The narrower plastic zone in laser welding is beneficial because it limits the region where strain age cracking can initiate, but the higher peak stress is a concern for the weld itself.
Integration with Manufacturing Practice
In my experience with high-temperature alloys, the practical question is always: which process gives us the best chance of a sound weld that will survive service? The paper's data suggests that laser welding, despite higher peak stress, may be preferable for thin sections because the smaller heat-affected zone limits the extent of the degraded region. However, I have seen cases where laser welding of thin superalloy components resulted in unacceptable distortion due to the steep thermal gradients. The key is to control the heat input and the clamping arrangement. The paper's finding that the transverse residual stress in the weld shows significant deviation between simulation and experiment is important. This is likely due to the difficulty of accurately modelling the weld pool geometry and the solidification process. In practice, this means that the simulation results for transverse stress should be treated with caution, and the weld design should account for the possibility of higher transverse stresses than predicted. For thin plates, transverse residual stress can cause buckling distortion, which is a common problem in laser welding of thin sections.
Key Questions and Reflections
The paper raises an important question about the applicability of the finite element model to real welding conditions. The authors used a heat source model that presumably approximates the actual heat input, but the validation showed discrepancies in the transverse stress field. This is not unusual, as transverse stress is highly sensitive to the details of the heat source distribution and the boundary conditions. In practice, I would recommend that the simulation be calibrated against actual weld tests for each specific joint configuration and thickness. The paper also highlights the difference in plastic strain between the two processes. The lower plastic strain in laser welding is beneficial for reducing the risk of strain age cracking, but it also means that the weld has less capacity to accommodate thermal expansion during service. This is a subtle point that is often overlooked. The weld must not only be free of cracks at the time of fabrication but must also survive the thermal cycles of service. The residual stress state influences the crack initiation and propagation behaviour, and the interaction between residual stress and service stress is complex.
Study Insights and Practical Recommendations
For the welding engineer working with IN738 or similar superalloys, the paper offers several practical insights. First, the choice between laser and TIG welding should be based on the specific requirements of the component, not on a general preference. If the component is thin and distortion is a concern, laser welding may be preferred despite the higher peak stress, because the smaller heat-affected zone reduces the overall distortion. If the component is thick or the weld is in a highly constrained location, TIG welding may be more appropriate because the broader heat input reduces the peak stress and the risk of hot cracking. Second, the residual stress state should be managed through post-weld heat treatment or stress relief. For IN738, the standard practice is to perform a solution treatment and aging after welding, which can relieve residual stresses and restore the precipitation hardening. However, the timing of the heat treatment is critical, as strain age cracking can occur during the heat treatment itself if the residual stresses are high. Third, the discrepancy between simulated and measured transverse stress highlights the need for validation. In production, I would recommend that the welding procedure qualification include residual stress measurement, using techniques such as hole drilling or X-ray diffraction, to verify that the actual residual stress state is acceptable.
Conclusion
This comparative study of laser and TIG welding of IN738 superalloy thin plates provides valuable data on the residual stress and plastic strain distributions for both processes. The key finding is that laser welding produces higher peak residual stress but a narrower stress zone and smaller plastic deformation zone, while TIG welding produces lower peak stress but a broader heat-affected zone. The practical implication is that process selection must consider the specific requirements of the component, balancing the risk of hot cracking, distortion, and strain age cracking. The discrepancy between simulated and measured transverse stress underscores the need for careful validation of finite element models against actual weld tests. For the production engineer, this paper reinforces the importance of understanding the residual stress state in welded superalloy components and managing it through appropriate process selection, heat treatment, and quality control. The ultimate goal is a weld that is free of defects at fabrication and remains reliable in service, and this paper provides useful guidance on achieving that goal.
Zhuojin Pipe Fitting Co., Ltd