Zhuojin Pipe Fitting Co., LtdNumerical Simulation of Temperature Field During TIG Welding of Invar Alloy and 45 Steel
Literature Overview
The paper by Xu Peiquan, Zhao Xiaohui, He Jianping, Xu Guoxiang, and Yu Shuizhi, published in the journal Welding Journal (Welding & Joining) in 2008 (Vol. 29, No. 6, pp. 37-40), presents a comprehensive numerical simulation study of the temperature field distribution during TIG welding of Invar alloy (Fe-36Ni) to 45 steel. Funded by the Shanghai Outstanding Youth Fund (06xpyq17), Shanghai Science and Technology Commission Key Project (061111034), and Shanghai Education Commission Key Discipline Project (J51402), this research was conducted at Shanghai University of Engineering Science (School of Materials Engineering) and Dalian Heavy Industries-Crane Group Co., Ltd. (Welding Technology Research Institute). The study employs ANSYS finite element software with a double ellipsoid heat source model to simulate temperature field evolution during welding of different plate thicknesses, achieving weak coupling between temperature and stress fields.
Theoretical Framework and Numerical Model
Double Ellipsoid Heat Source Model
The double ellipsoid heat source model, originally proposed by Goldak, Akerman, and Butt, is the cornerstone of this numerical simulation. This model represents the welding arc as two half-ellipsoids: a forward (pre-heat) ellipsoid and a rearward (cooling) ellipsoid, each with distinct energy distribution characteristics. The heat flux density is expressed as:
q(x, y, z) = (6√3 · Q) / (a · b · c · π^(3/2)) · exp(-3x²/(a²) - 3y²/(b²) - 3z²/(c²))
where Q is the effective heat input, and a, b, c are the ellipsoid semi-axes in the longitudinal, transverse, and depth directions respectively. The forward ellipsoid has a smaller depth parameter (c₁) representing rapid heat deposition, while the rearward ellipsoid has a larger depth parameter (c₂) representing slower heat dissipation. This model accurately captures the asymmetric heat distribution characteristic of TIG welding, where the molten pool is elongated in the direction of travel.
Weak Coupling of Temperature and Stress Fields
The authors implement a weak coupling approach between the temperature field and stress field simulations. In this approach, the temperature field is solved first, and the resulting temperature distribution is used as a thermal load for the subsequent stress analysis. This sequential approach is computationally efficient and appropriate for the elastic-plastic analysis of welding residual stresses, where the thermal expansion/contraction is the primary driver of mechanical deformation. The weak coupling assumption is valid when the mechanical deformation does not significantly affect the thermal conductivity or heat capacity of the material—a reasonable assumption for the temperature ranges encountered in TIG welding.
Material Properties and Boundary Conditions
The simulation requires accurate material property data for both Invar alloy and 45 steel across the temperature range from ambient to solidus temperature. Key properties include:
| Property | Invar Alloy (Fe-36Ni) | 45 Steel | Temperature Dependence |
|---|---|---|---|
| Thermal conductivity (W/m·K) | 14-17 | 45-50 | Decreases with temperature |
| Specific heat (J/kg·K) | 500-550 | 460-500 | Increases with temperature |
| Thermal expansion (×10⁻⁶/K) | 1.2-1.5 | 12-13 | Strongly temperature-dependent |
| Density (kg/m³) | 8080 | 7850 | Slight decrease with temperature |
| Melting point (°C) | 1425-1470 | 1500-1520 | — |
| Young's modulus (GPa) | 150-170 | 205 | Decreases above 0.4Tm |
The extremely low coefficient of thermal expansion of Invar alloy (approximately 1/10 that of carbon steel) is the defining characteristic that makes this material valuable for precision engineering applications but also creates unique challenges during welding, including differential shrinkage at the joint interface and potential cracking due to thermal mismatch.
Simulation Results and Experimental Validation
Optimal Process Parameters
The authors identify the following process parameters as optimal for achieving full penetration welding of 1.88 mm thick Invar alloy plate:
| Parameter | Value | Unit |
|---|---|---|
| Welding current | 132 | A |
| Arc voltage | 17.1 | V |
| Travel speed | 4 | mm/s |
| Plate thickness | 1.88 | mm |
| Polarity | DCEN | — |
| Shielding gas | Argon | — |
| Gas flow rate | 12-15 | L/min |
These parameters represent a relatively low heat input configuration (approximately 0.57 kJ/mm), which is consistent with the need to minimize thermal distortion when welding Invar alloy. The low travel speed combined with moderate current ensures adequate penetration while limiting the thermal cycle severity.
Temperature Field Distribution
The simulation results reveal several important characteristics of the temperature field during TIG welding of Invar alloy:
- Peak temperature: The maximum temperature in the molten pool reaches approximately 1800-2000°C, well above the melting point of both materials, ensuring complete fusion at the joint interface.
- Thermal gradient: The thermal gradient at the fusion boundary is steep (approximately 50-100°C/mm), creating a narrow heat-affected zone (HAZ) that is critical for minimizing microstructural changes in the base metal.
- Asymmetric distribution: The temperature field exhibits pronounced asymmetry in the direction of travel, with higher temperatures ahead of the arc center (forward ellipsoid effect) and rapid cooling behind the arc.
- Interface effects: The temperature distribution at the Invar-45 steel interface shows a slight asymmetry due to the different thermal conductivities of the two materials, with heat preferentially flowing into the 45 steel side.
Experimental Validation
The authors validate the numerical simulation through welding experiments where deformation at specific locations was measured and compared with simulation predictions. The good agreement between simulation and experimental results confirms the accuracy of the numerical model and the appropriateness of the double ellipsoid heat source parameters. This validation is essential for establishing confidence in the simulation methodology and its applicability to other welding scenarios.
Engineering Practice Implications for Dissimilar Metal Welding
The welding of Invar alloy to carbon steel is a challenging dissimilar metal welding problem that arises in several engineering applications:
- Precision instrumentation: Invar's near-zero thermal expansion is exploited in precision devices, but structural support often requires carbon steel components.
- Aerospace applications: Thermal management systems may combine Invar components with conventional steel structures.
- Thermal cycling environments: Equipment subject to extreme temperature variations may use Invar for dimensional stability while connecting to steel piping or frames.
From a quality control perspective, several FMEA (Failure Mode and Effects Analysis) considerations are critical:
| Failure Mode | Cause | Effect | Detection Method | Prevention |
|---|---|---|---|---|
| Cracking at fusion boundary | Thermal mismatch, high cooling rate | Joint failure | RT/UT inspection | Preheating, low heat input |
| Excessive distortion | Differential thermal expansion | Dimensional inaccuracy | CMM measurement | Fixturing, symmetric welding |
| Porosity | Gas entrapment, incomplete shielding | Reduced joint strength | RT/UT inspection | Proper gas flow, clean surfaces |
| Lack of penetration | Insufficient heat input | Incomplete fusion | UT/RT inspection | Parameter optimization |
| Microstructural degradation | Overheating in HAZ | Reduced toughness | Microhardness mapping | Controlled thermal cycles |
The numerical simulation approach demonstrated in this paper provides a powerful tool for predicting and mitigating these failure modes before physical welding trials are conducted, reducing development time and material costs.
Key Technical Insights and Reflections
Several aspects of this research deserve particular attention from a practical engineering standpoint. The use of the double ellipsoid heat source model represents a mature and well-validated approach for welding thermal analysis, but its accuracy depends critically on the proper calibration of the ellipsoid parameters. The authors' approach of calibrating these parameters against experimental results is the correct methodology, as the heat source geometry is inherently process-dependent and cannot be assumed universal.
The weak coupling approach between temperature and stress fields is a pragmatic choice that balances computational efficiency with analytical accuracy. For the specific case of Invar alloy welding, where thermal distortion is a primary concern, this approach provides sufficient accuracy for process optimization while remaining computationally tractable. However, for applications where large plastic deformation is expected, a fully coupled thermo-mechanical analysis might be necessary.
A significant insight from this research is the demonstration that numerical simulation can effectively predict welding outcomes for dissimilar metal joints where experimental investigation is expensive or impractical. The ability to systematically vary process parameters in simulation and identify optimal configurations before physical trials represents a significant productivity improvement in welding process development.
Study Implications for Steel Pipe and Fitting Manufacturing
For engineers in steel pipe and fitting manufacturing, this research offers several transferable insights:
- Process development efficiency: Numerical simulation of temperature fields can accelerate the development of welding procedures for dissimilar metal joints encountered in piping systems (e.g., carbon steel to stainless steel, carbon steel to nickel alloys).
- Distortion prediction: The methodology can be adapted to predict welding distortion in large-diameter pipe spools and complex fitting geometries, enabling the design of appropriate fixturing and post-weld straightening operations.
- Heat-affected zone characterization: Temperature field simulation provides quantitative data on thermal cycles experienced by the HAZ, supporting predictions of microstructural evolution and mechanical property changes.
- Welding procedure specification (WPS) optimization: Simulation results can inform the development of WPS parameters that balance penetration, distortion control, and mechanical performance.
The temperature field simulation methodology presented in this paper is directly applicable to the analysis of TIG welding processes used in the production of thin-walled pipe fittings, instrumentation tubing, and precision components where thermal distortion control is critical.
Conclusion
The research by Xu Peiquan and colleagues demonstrates the effectiveness of finite element numerical simulation in predicting temperature field distribution during TIG welding of Invar alloy to 45 steel. The double ellipsoid heat source model, implemented in ANSYS with weak coupling of thermal and stress fields, provides accurate predictions of temperature distribution that are validated by experimental measurements. The identification of optimal process parameters (132 A, 17.1 V, 4 mm/s for 1.88 mm plate) provides a practical starting point for welding procedure development. For steel pipe and fitting manufacturers, the key takeaway is that numerical simulation is a powerful tool for welding process development, distortion prediction, and quality assurance, particularly for challenging dissimilar metal welding applications where experimental investigation alone would be prohibitively expensive and time-consuming. The methodology presented in this paper should be integrated into the welding engineering toolkit as a standard practice for process optimization and quality prediction.