Numerical Simulation and Experimental Verification of TIG Welding Temperature Field for 0Cr18Ni10Ti Stainless Steel Source Shell

Literature Overview and Research Context

This 2015 study by Luo Hongyi, Tang Xian, and Luo Zhifu from the China Institute of Atomic Energy addresses a highly specialized and safety-critical welding application: the TIG welding of 0Cr18Ni10Ti (equivalent to UNS S31803 or similar stabilized austenitic stainless steel) source shells for nuclear fuel fabrication. The China Institute of Atomic Energy is a premier research institution for nuclear materials and fuel fabrication in China, and the work reflects the rigorous engineering standards required for nuclear-grade components. The study was published in the journal of Atomic Energy Science and Technology, underscoring its relevance to the nuclear industry.

The welding of nuclear fuel source shells is among the most demanding welding applications in practice. These shells must exhibit excellent dimensional accuracy, surface finish, and metallurgical integrity, as any defect or deviation could compromise the performance and safety of the nuclear fuel assembly. The temperature field during welding is a critical parameter that governs the solidification behavior, microstructure evolution, residual stress distribution, and final mechanical properties of the weld. Numerical simulation of the welding temperature field provides valuable insights that are difficult or impossible to obtain through experimental measurement alone, particularly for thin-walled components where thermocouple placement is impractical.

Material Characteristics and Welding Challenges

0Cr18Ni10Ti is a titanium-stabilized austenitic stainless steel containing approximately 18 percent chromium, 10 percent nickel, and a small amount of titanium. The titanium addition serves to tie up carbon as titanium carbide, preventing chromium carbide precipitation at grain boundaries and thereby improving resistance to intergranular corrosion. This alloy is commonly used in nuclear applications where resistance to corrosion and radiation-induced degradation is required.

The welding of thin-walled 0Cr18Ni10Ti source shells presents several specific challenges. First, the low thermal conductivity and high thermal expansion coefficient of austenitic stainless steels lead to significant thermal distortion during welding, which must be controlled to maintain dimensional accuracy. Second, the thin wall thickness limits the available heat input, as excessive heat input can cause burn-through, while insufficient heat input can result in incomplete fusion. Third, the narrow heat-affected zone (HAZ) in thin sections leaves little margin for error in the welding parameters, and any deviation from optimal conditions can result in unacceptable microstructural changes or residual stresses.

Numerical Simulation Methodology

The numerical simulation of the welding temperature field typically employs finite element analysis (FEA) using a moving heat source model. The most common approaches are the double-ellipsoidal heat source model proposed by Goldak and the conical heat source model, both of which can represent the asymmetric heat distribution characteristic of TIG welding. The simulation is performed in a thermomechanical coupling framework, where the temperature field is solved first and then used as input for the stress-strain analysis.

The moving heat source model is essential for simulating the welding process because the heat input moves with the welding speed, creating a non-steady-state temperature field. The Goldak double-ellipsoidal model divides the heat source into a front zone (keyhole) and a rear zone (weld pool tail), each with different heat distribution characteristics. The front zone has a smaller volume and higher heat intensity, while the rear zone has a larger volume and lower heat intensity. This asymmetry is critical for accurately predicting the weld pool shape and the temperature distribution in the weld and HAZ.

Experimental Verification and Comparison

The experimental verification of the numerical simulation results is essential for establishing the credibility of the simulation model. The experimental work typically involves measuring the temperature distribution on the surface of the weldment using infrared thermography or embedded thermocouples, and comparing the measured temperatures with the simulated temperatures at corresponding locations.

The comparison between simulated and experimental temperature fields typically shows good agreement in the overall trend, with deviations of 10–30 percent in peak temperatures and larger deviations in the temperature gradient regions. The deviations are primarily attributed to simplifications in the heat source model, uncertainties in boundary conditions (particularly the convective heat transfer coefficient and emissivity), and the difficulty of accurately representing the material properties at high temperatures.

Thermal Analysis and Residual Stress Implications

The temperature field during welding directly determines the residual stress distribution in the weldment. As the weld metal cools from the liquidus temperature to room temperature, it contracts. However, this contraction is constrained by the surrounding cooler material, which has not undergone the same degree of thermal expansion and contraction. The result is a complex residual stress field that includes tensile stresses in the weld and HAZ, balanced by compressive stresses in the far-field material.

For nuclear fuel source shells, the residual stress distribution is particularly important because it affects the dimensional stability of the shell during subsequent processing and service. Excessive residual stresses can cause distortion during machining or heat treatment, leading to dimensional deviations that may compromise the fit and function of the fuel assembly. Post-weld stress relief heat treatment is often employed to reduce residual stresses, but this introduces additional thermal cycles that must be carefully controlled to avoid sensitization or other adverse microstructural changes.

Engineering Practice Integration

In the context of nuclear fuel fabrication, the numerical simulation results presented in this study have direct practical value. The simulation can be used to optimize welding parameters for specific shell geometries and thicknesses, predict distortion patterns, and design appropriate fixturing and support strategies to control deformation. The simulation can also be used to evaluate the effects of different welding sequences on the residual stress distribution, which is particularly important for multi-pass welding of large-diameter shells.

For quality assurance purposes, the simulation results can be used to establish acceptance criteria for welding parameters and to define the boundaries of the qualified welding procedure. The simulation can also be used to perform what-if analyses, evaluating the effects of parameter variations on the temperature field and residual stress distribution without the need for expensive and time-consuming experimental trials.

Key Questions and Reflections

A significant question that arises from this study is the extent to which the simulation results can be extrapolated to different geometries, thicknesses, and welding conditions. While the simulation model is validated for a specific shell geometry and set of welding parameters, the applicability of the model to other configurations must be established through additional validation studies. Additionally, the simulation of the microstructural evolution during welding—such as grain growth, phase transformation, and precipitation—requires more sophisticated models that go beyond the thermal analysis presented here.

Another important consideration is the coupling of the thermal simulation with the mechanical simulation. While the temperature field is the primary driver of the residual stress distribution, the mechanical response of the material—including plastic deformation, strain hardening, and creep—also plays a significant role in determining the final residual stress state. A fully coupled thermomechanical simulation would provide more accurate predictions of the residual stress distribution, but at the cost of increased computational complexity.

Study Insights and Implications

The most valuable contribution of this study is the demonstration that numerical simulation can provide reliable predictions of the welding temperature field for thin-walled stainless steel components, with experimental validation confirming the accuracy of the simulation model. For nuclear fuel fabrication, where the consequences of welding defects are severe and the regulatory requirements are stringent, the use of numerical simulation as a design and optimization tool is not merely beneficial but essential. The study also highlights the importance of experimental validation in establishing the credibility of simulation models, a principle that applies broadly to all engineering applications. For engineers involved in the fabrication of nuclear components and other critical pressure vessels, this work demonstrates that the combination of numerical simulation and experimental verification can provide the confidence needed to optimize welding processes while maintaining the highest standards of quality and safety.