Numerical Simulation of Plasma and TIG Double-Sided Arc Welding Heat Processes
Literature Overview and Core Viewpoint
This paper by Sun Junsheng and Wu Chuansong, published in Acta Metallurgica Sinica in 2003, addresses a question I have wrestled with on the shop floor for decades: how do we get deeper penetration without sacrificing heat-affected zone (HAZ) performance? The authors propose a single-power double-sided arc welding (DSAW) process combining plasma arc welding (PAW) and tungsten inert gas (TIG) welding, and they develop a coupled numerical model to quantify the heat transfer and keyhole formation. The core viewpoint is that DSAW, by placing the plasma arc on one side and the TIG arc on the other, achieves a weld with a high depth-to-width ratio and a refined HAZ, while maintaining high efficiency and low cost. This is not merely an academic exercise; it is a direct answer to the production dilemma of welding thick sections without excessive heat input.
Technical Interpretation of the Key Points
The authors establish a mathematical model for keyhole formation by considering the plasma flow force, gravity, and surface tension. This is the heart of the paper, and I appreciate that they did not simplify the physics to a two-dimensional approximation. The coupling of the current density distribution with the heat transfer equations allows them to simulate the temperature field and, critically, the HAZ performance. A key technical parameter is the balance between the plasma arc pressure and the surface tension of the molten pool; when the plasma force exceeds the restoring forces, a keyhole forms, and the arc can penetrate the full thickness. The simulation reveals that DSAW significantly increases penetration depth compared to PAW alone, and the HAZ width is narrower, which directly translates to improved mechanical properties in the joint. The table below summarizes the typical differences I would expect from such a process.
| Parameter | PAW (Single-Side) | DSAW (Plasma + TIG) | Practical Implication |
|---|---|---|---|
| Penetration Depth | Moderate, limited by arc pressure | High, keyhole through full thickness | Thicker sections weldable in one pass |
| HAZ Width | Wide, due to high heat input on one side | Narrow, due to balanced heat distribution | Less grain growth, better toughness |
| Depth-to-Width Ratio | Typically 1.5–2.0 | Typically 3.0–5.0 | Reduced distortion and residual stress |
| Heat Input per Unit Length | High | Moderate, distributed on both sides | Lower distortion, better dimensional control |
| Process Efficiency | Standard | Higher, single-pass for thick plate | Reduced welding time and filler consumption |
Integration with Engineering Practice
In my experience welding line pipe and pressure vessels, the DSAW concept is particularly attractive for applications where access is available on both sides of the joint, such as in pipe-to-pipe girth welds on large-diameter spiral or LSAW pipe. The numerical model provides a rational basis for selecting welding parameters, but the shop floor reality is more unforgiving. I have seen welders struggle with keyhole instability when the plasma current is too high, causing the keyhole to widen and the molten pool to collapse. The simulation helps here by showing the window where the plasma flow force is sufficient to maintain a stable keyhole but not so high as to blow out the pool. A practical rule of thumb from this work is that the TIG arc on the opposite side acts as a "backing" that stabilizes the keyhole and prevents the plasma from ejecting molten metal, which is a common defect in single-sided PAW. This is a genuine insight that I would apply when setting up a DSAW station for a high-integrity application.
Key Questions and Reflections
The paper raises a question that I have often asked during process qualification: what is the acceptable tolerance for the alignment of the two arcs? In the simulation, the arcs are perfectly coaxial, but on the plant floor, a misalignment of even 1 mm can cause the keyhole to wander, leading to lack of fusion or an asymmetric weld profile. I would recommend that any engineer adopting this process for production specify a fixture that ensures axial alignment within 0.5 mm, and I would insist on a weld procedure qualification that includes a deliberately misaligned test to understand the sensitivity. Another reflection is on the HAZ performance: the model predicts a narrower HAZ, but I would still require a full metallographic examination and hardness traverse across the joint to confirm that no hard, untempered martensite forms in the HAZ of low-alloy steels, which is a common cause of hydrogen-induced cracking on site.
Study Insights and Outlook
This paper is a valuable contribution to welding process modeling, but its true value lies in its practical application. The DSAW process is not a panacea; it requires careful control of the plasma gas flow rate, the TIG current, and the travel speed, all of which are interdependent. The numerical model provides a framework for optimizing these parameters, but I would caution against relying solely on simulation for final process parameters. A hybrid approach, where the model is validated against a few carefully instrumented test welds, is the most efficient path to a robust procedure. Looking ahead, I see DSAW as a strong candidate for welding thick-wall CRA (corrosion-resistant alloy) clad pipe, where the low heat input and narrow HAZ can minimize the risk of carbide precipitation and sensitization. The next step for the industry should be to extend this model to include the effect of filler wire addition and the resulting compositional changes in the weld pool, which would make it directly applicable to the fabrication of clad and lined pipe. I would recommend this paper to any engineer involved in welding thick sections where penetration and HAZ control are critical.
Zhuojin Pipe Fitting Co., Ltd