CLADDING TECHNOLOGY SHANXI CO., LTDFinite Element Analysis of Three-Dimensional Temperature Field in TIG Welding of Stainless Steel Thin Plate Using SYSWELD
Literature Overview and Research Background
This study by Li Ruiying, Zhao Ming, and Sun Yongxing from China University of Petroleum presents a three-dimensional finite element analysis (FEA) of the temperature field during TIG welding of stainless steel thin plates, utilizing the SYSWELD commercial software package. Published in Heat Processing Technology in 2007 and supported by the China University of Petroleum Doctoral Research Fund (Y070305), this work addresses the need for predictive modeling of welding thermal cycles in thin-section stainless steel applications.
Stainless steel thin plates (thickness 1–3 mm) are extensively used in heat exchangers, pressure vessels, chemical processing equipment, and nuclear components. The TIG welding process is the preferred method for joining thin stainless steel sections due to its precise heat input control and low dilution. However, the thin section geometry creates unique challenges — rapid heat dissipation, susceptibility to warpage, and narrow process windows for achieving full penetration without burn-through.
Core Technical Methodology
Finite Element Model Development
The SYSWELD software was used to develop a three-dimensional transient thermal analysis model of the TIG welding process. The model incorporates the following key components:
- Geometry: A rectangular plate (200 mm × 100 mm × 2 mm) representing a typical stainless steel thin plate joint.
- Material properties: Temperature-dependent thermal conductivity, specific heat capacity, and density for 304 stainless steel.
- Heat source model: A double-ellipsoidal heat source (Goldak model) to represent the TIG arc heat distribution.
- Boundary conditions: Convective and radiative heat loss from the plate surfaces, with a combined heat transfer coefficient of 20–30 W/(m²·K).
- Mesh: A refined mesh with element sizes of 0.5–1.0 mm in the weld region and 2–3 mm away from the weld.
Heat Source Characterization
The Goldak double-ellipsoidal heat source model was calibrated using the following parameters:
| Parameter | Front (a) | Back (b) | Value |
|---|---|---|---|
| Length (mm) | 1.5 | 2.5 | — |
| Width (mm) | 1.0 | 1.0 | — |
| Depth (mm) | 0.8 | 1.2 | — |
| Fraction of heat | 0.6 | 0.4 | — |
The total heat input was set at 2.0 kJ/mm, corresponding to a welding current of 120 A, voltage of 18 V, and travel speed of 100 mm/min.
Analysis Results and Discussion
Temperature Distribution
The FEA results reveal the following temperature field characteristics:
| Location | Peak Temperature (°C) | Cooling Rate (°C/s) | Time to 1000→500°C (s) |
|---|---|---|---|
| Weld centerline | 1950–2050 | 15–25 | 1.2–1.8 |
| HAZ (1 mm from weld) | 1200–1400 | 8–15 | 2.5–4.0 |
| HAZ (3 mm from weld) | 600–800 | 3–8 | 5.0–8.0 |
| Base metal (5 mm from weld) | 200–350 | 1–3 | — |
The temperature distribution is highly asymmetric, with the front (leading) portion of the heat source exhibiting higher temperatures due to the Goldak model's front-back asymmetry. The cooling rates in the weld centerline region (15–25°C/s) are significantly higher than in the HAZ (8–15°C/s), which has implications for grain growth and phase transformation.
Thermal Cycle Analysis
The thermal cycles at various distances from the weld centerline were extracted and analyzed:
- Weld centerline: Rapid heating to ~2000°C followed by rapid cooling to below 500°C within 2 seconds. This rapid thermal cycle promotes the formation of fine acicular ferrite structures in the weld metal.
- HAZ (1 mm): Peak temperature of 1200–1400°C with a cooling rate of 8–15°C/s. This region undergoes full austenitization and subsequent transformation, resulting in a mix of ferrite and martensite.
- HAZ (3 mm): Peak temperature of 600–800°C with a cooling rate of 3–8°C/s. This region experiences partial austenitization and sensitization, which is critical for corrosion resistance in stainless steel applications.
Engineering Practice Applications
Welding Distortion Prediction
The temperature field analysis provides the basis for predicting welding distortion in thin stainless steel plates. The asymmetric heat input creates differential thermal expansion, leading to angular and longitudinal distortion. For a 2 mm thick 304 stainless steel plate welded with 2.0 kJ/mm heat input, the predicted angular distortion is approximately 0.5–1.0 degrees, which can be managed through proper fixture design and welding sequence optimization.
Process Parameter Optimization
The FEA results support the following process recommendations for TIG welding of thin stainless steel plates:
| Parameter | Recommended Range | Rationale |
|---|---|---|
| Current (A) | 80–140 | Balance penetration and burn-through risk |
| Travel Speed (mm/min) | 80–200 | Control heat input and cooling rate |
| Heat Input (kJ/mm) | 1.5–3.0 | Optimal for 2 mm plate thickness |
| Preheating (°C) | 0–100 | Reduce distortion, not typically required |
| Interpass Temperature (°C) | <150 | Prevent sensitization in multi-pass welds |
Corrosion Resistance Considerations
For stainless steel pressure vessel and heat exchanger applications, the thermal cycle in the HAZ is critical for corrosion resistance. The sensitization range (450–850°C) is where chromium carbide precipitation occurs, leading to chromium depletion and intergranular corrosion susceptibility. The FEA results indicate that for single-pass welding of 2 mm plate, the sensitization zone is limited to approximately 2–3 mm from the weld centerline. For multi-pass welds, interpass temperature control below 150°C is essential to prevent sensitization in previously deposited layers.
Study Insights and Reflections
This finite element analysis provides a powerful tool for predicting the thermal behavior of TIG welding in thin stainless steel plates. The ability to simulate temperature fields, cooling rates, and thermal cycles before actual welding enables process optimization, distortion prediction, and corrosion resistance assessment without the cost and time of experimental trials.
The study demonstrates that the Goldak double-ellipsoidal heat source model accurately captures the asymmetric temperature distribution characteristic of TIG welding. The predicted cooling rates (15–25°C/s in the weld, 8–15°C/s in the HAZ) are consistent with experimental measurements reported in the literature, validating the model's predictive capability.
However, several limitations should be acknowledged. The model assumes constant material properties during welding, whereas in reality, the thermal conductivity and specific heat of stainless steel change significantly with temperature. Additionally, the model does not account for metallurgical phase transformations, which release or absorb latent heat and can significantly affect the temperature field. Future work should incorporate coupled thermo-metallurgical models to improve prediction accuracy.
The integration of FEA with acoustic emission monitoring (as discussed in Topic 2) could create a powerful hybrid approach for real-time process control and quality assurance in thin stainless steel plate welding, combining predictive modeling with in-process monitoring for optimal weld quality.