CLADDING TECHNOLOGY SHANXI CO., LTDNumerical Simulation of Temperature and Flow Fields in Aluminum-Steel TIG Welding-Brazing
Literature Overview
This research by Song Yang and colleagues from Dalian University of Technology, published in Welding in 2014, presents a numerical simulation study of the temperature field and flow field during TIG welding-brazing of aluminum to steel. Funded by the National Natural Science Foundation of China (50904012/E041607) and the Liaoning Provincial Natural Science Foundation (20092152), this work addresses the fundamental challenge of understanding the complex thermal and fluid dynamics that govern the quality of dissimilar metal joints.
Simulation Methodology
The numerical model employs finite element analysis to solve the coupled heat transfer and fluid flow equations in the weld pool. The governing equations include the energy equation with enthalpy-temperature formulation to account for phase change, and the Navier-Stokes equations with Boussinesq approximation for natural convection. The arc heat source is modeled as a double-ellipsoidal distribution, which captures the asymmetric temperature profile in the direction of travel.
Key assumptions and boundary conditions include:
- Heat transfer at the pool surface accounts for radiation, convection, and evaporative cooling
- The steel side is modeled as a solid with temperature-dependent thermal properties
- The aluminum side is modeled with phase change from solid to liquid
- Surface tension gradient (Marangoni convection) is included as a driving force for pool flow
- Gravity-induced natural convection is considered
Thermal Field Results
The simulation reveals that the peak temperature at the pool center reaches approximately 1450-1550°C, which exceeds the melting point of aluminum (660°C) but remains below the solidus of austenitic stainless steel (approximately 1400°C). This confirms the feasibility of the welding-brazing concept—the aluminum melts and flows onto the steel surface while the steel remains solid.
The temperature distribution shows a pronounced asymmetry. The aluminum side experiences a wider thermal gradient due to its higher thermal conductivity (approximately 200 W/m·K versus 15-20 W/m·K for stainless steel). The isotherm lines are compressed on the aluminum side and elongated on the steel side, creating a characteristic asymmetric weld pool shape.
The peak temperature in the steel HAZ, located at the interface, reaches approximately 500-700°C depending on the process parameters. This temperature range is well below the Ac1 of austenitic stainless steel, confirming that no phase transformation occurs in the steel HAZ. However, temperatures above 400°C in the steel HAZ can lead to temper embrittlement in precipitation-hardened stainless steels, which is a consideration for certain grades.
Flow Field Analysis
The flow field in the aluminum weld pool is driven primarily by Marangoni convection caused by surface tension gradients. The surface tension of aluminum decreases with temperature, creating an outward flow from the pool center toward the cooler regions. This outward flow is opposed by buoyancy-driven natural convection, which drives flow from the hot bottom toward the cooler top surface.
The interaction between these two flow mechanisms creates a complex circulation pattern:
- Near the pool surface: outward flow driven by negative surface tension gradient (dγ/dT < 0)
- In the pool interior: inward return flow driven by buoyancy
- At the pool boundaries: complex three-dimensional flow patterns influenced by the temperature gradient
The flow velocity in the aluminum pool reaches maximum values of 0.5-1.2 m/s near the pool surface, which is sufficient to transport heat and dissolved elements throughout the pool. The steel side, being solid, does not participate in convective flow, but the temperature gradient at the interface drives diffusion of aluminum into the steel surface layer.
Parameter Study and Process Optimization
The simulation was used to investigate the effects of key process parameters on the thermal and flow fields:
| Parameter | Effect on Peak Temperature | Effect on Pool Geometry | Effect on Steel HAZ Temperature |
|---|---|---|---|
| Arc current increase | Increases peak temperature by 50-100°C | Widens and deepens pool | Increases interface temperature by 50-80°C |
| Travel speed increase | Decreases peak temperature by 80-150°C | Narrows and shallows pool | Decreases interface temperature by 60-100°C |
| Shielding gas flow increase | Minimal effect on peak temperature | Slightly narrows pool | Minimal effect |
| Joint gap increase | Increases peak temperature slightly | Deepens pool | Increases interface temperature |
The simulation results provide quantitative guidance for process optimization. For example, increasing the travel speed from 250 to 350 mm/min reduces the steel HAZ peak temperature by approximately 80°C, which is beneficial for minimizing thermal distortion and intermetallic formation. However, excessive travel speed reduces pool wetting and may lead to incomplete joint formation.
Engineering Practice Integration
The numerical simulation results complement experimental investigations by providing insights into regions that are difficult to measure experimentally, such as the instantaneous temperature distribution within the weld pool and the subsurface flow patterns. For welding procedure development, the simulation can be used as a screening tool to identify promising parameter combinations before expensive experimental trials.
In pressure vessel fabrication, where aluminum-to-steel joints may be used in heat exchanger tubesheets or in lightweight structural components, the simulation provides a basis for predicting the thermal history of the joint. This thermal history directly influences the microstructure and mechanical properties of the interface region. The simulation can also be extended to predict residual stress distributions, which is critical for assessing the long-term integrity of the joint under operational loads.
I find particularly valuable the insight that the Marangoni convection, rather than buoyancy, dominates the pool flow in aluminum welding-brazing. This has direct implications for filler metal placement and arc positioning. Positioning the arc slightly ahead of the filler wire, in the direction of travel, can enhance the outward flow and improve wetting of the steel surface. Conversely, positioning the arc behind the filler wire may lead to excessive inward flow and poor joint coverage.
The study also highlights the importance of surface preparation. Oxide films on the steel surface act as a barrier to aluminum wetting, and the simulation assumes a clean interface. In practice, mechanical cleaning or chemical treatment of the steel surface is essential to achieve the predicted thermal and flow conditions. This practical consideration is often overlooked in purely numerical studies but is critical for successful production welding.