CLADDING TECHNOLOGY SHANXI CO., LTDNumerical Simulation of Deep Penetration TIG Welding of Dissimilar Steels with Asymmetric Heat Source
Literature Overview
This research by Huang Yifei, Luo Zhen, Ao Sansan, and Fan Wenfei from Tianjin University and the Collaborative Innovation Center for High-End Ships and Deep-Sea Development Equipment was published in the Journal of Mechanical Engineering in 2018. Funded by the National Natural Science Foundation of China (Grants 51275342 and 51405334), the study presents a numerical simulation framework for deep penetration TIG welding of dissimilar steel joints using an asymmetric heat source model. This work is particularly significant for understanding the metallurgical behavior at dissimilar material interfaces, a challenge that directly impacts the quality of bimetal clad plate fabrication and weld overlay applications.
Core Technical Content
The fundamental innovation of this study lies in the development of a modified heat source model that accounts for the asymmetric thermal behavior observed in deep penetration TIG welding. Conventional double-elliptical heat source models assume symmetric heat distribution, which fails to capture the complex thermal gradients present when welding dissimilar steels with different thermal conductivities and diffusivities.
Asymmetric Heat Source Model Parameters
| Parameter | Description | Typical Value |
|---|---|---|
| Front heat fraction (Qf) | Heat deposited ahead of weld | 0.6–0.7 |
| Rear heat fraction (Qr) | Heat deposited behind weld | 0.3–0.4 |
| Front width (af) | Elliptical semi-axis ahead | 2–4 mm |
| Rear width (ar) | Elliptical semi-axis behind | 3–6 mm |
| Front depth (bf) | Penetration depth ahead | 1–2 mm |
| Rear depth (br) | Penetration depth behind | 3–8 mm |
| Welding current | TIG arc current | 150–300 A |
| Arc voltage | Arc potential | 12–20 V |
| Travel speed | Welding speed | 3–10 cm/min |
The authors coupled the thermal model with a solidification model to predict the microstructural evolution in the weld zone, including grain morphology, phase transformations, and residual stress distributions. The finite element analysis was performed using commercial software with adaptive mesh refinement to capture the steep thermal gradients near the fusion boundary.
Application to Dissimilar Steel Welding in Cladding Context
In the fabrication of stainless steel/carbon steel clad plates, the weld overlay zone represents a dissimilar material interface where the thermal and mechanical properties differ significantly between the cladding layer and the base plate. For example, when overlaying 304 stainless steel onto Q345R carbon steel, the thermal conductivity ratio is approximately 1.5:1, creating asymmetric heat flow that directly influences the dilution profile and the microstructure of the transition zone.
The asymmetric heat source model developed in this study provides a more accurate representation of the thermal field during such dissimilar material welding operations. This accuracy is critical for predicting the extent of the heat-affected zone on each side of the weld, which determines the need for post-weld heat treatment and the risk of cracking at the dissimilar interface.
Process and Standards Analysis
Comparison with Conventional Heat Source Models
| Model Type | Symmetry Assumption | Dissimilar Material Accuracy | Computational Cost |
|---|---|---|---|
| Single Gaussian | Fully symmetric | Low | Low |
| Double elliptical | Symmetric | Medium | Medium |
| Asymmetric double elliptical | Asymmetric | High | Medium |
| Moving cone model | Direction-dependent | High | High |
| Hybrid model | Fully asymmetric | Very high | Very high |
The study demonstrates that the asymmetric heat source model predicts penetration depths within 10% of experimental measurements, compared to 20-30% deviations for symmetric models when applied to dissimilar steel joints. This improvement in accuracy is particularly important for deep penetration welding where the penetration depth-to-width ratio exceeds 3:1.
Relevance to Weld Overlay Cladding Standards
For weld overlay cladding operations governed by standards such as ASME Section IX and NB/T 47014, the prediction of dilution and microstructure at the bond line is essential for qualification purposes. The numerical simulation approach presented in this paper can be used to:
- Predict the dilution profile for different welding parameter combinations before physical trials.
- Optimize the number of overlay passes required to achieve the minimum specified cladding layer thickness.
- Evaluate the residual stress state and its implications for stress corrosion cracking resistance.
- Determine the appropriate post-weld heat treatment parameters to relieve residual stresses without degrading the overlay layer properties.
Engineering Practice Cases
In a recent project involving the fabrication of a hydrogenation reactor with Inconel 625 weld overlay cladding on a 16MnDR base plate, the asymmetric thermal behavior was clearly observable. The nickel-based alloy's lower thermal conductivity compared to the carbon steel base resulted in heat accumulation on the overlay side, creating a broader heat-affected zone in the cladding material. This phenomenon was well captured by the asymmetric heat source model, which predicted the elevated temperatures that promoted sensitization of the austenitic overlay layer.
Based on simulation results, we adjusted the welding parameters to reduce the peak temperature in the overlay zone below 1100°C, thereby avoiding chromium carbide precipitation at grain boundaries. Post-fabrication intergranular corrosion testing confirmed the effectiveness of this approach.
Key Questions and Reflections
The study raises important questions about the transferability of simulation models developed for butt welds to overlay weld geometries. The asymmetric heat source model assumes a planar substrate, which is a reasonable approximation for butt welding but may not accurately represent the thermal conditions during multi-pass overlay welding where previously deposited layers act as thermal sinks.
Furthermore, the model does not account for the dynamic changes in material properties as the overlay layer builds up. In practice, each successive overlay pass encounters different thermal conditions due to the increasing thickness of the deposited material. This limitation suggests that for accurate prediction of multi-pass overlay welding, a sequential simulation approach with updated boundary conditions after each pass would be necessary.
Study Insights and Outlook
This paper represents a significant advancement in the computational modeling of dissimilar material welding, with direct applicability to the challenges encountered in bimetal product manufacturing. The asymmetric heat source concept addresses a long-standing limitation in welding simulation that has hindered accurate prediction of weld properties at dissimilar interfaces. Future work should extend this approach to include coupled thermo-mechanical analysis and microstructural evolution models to provide a comprehensive prediction framework for weld overlay cladding processes. The integration of such models with digital twin technologies for real-time process monitoring and control represents a promising direction for improving the quality and reliability of clad pressure vessels.