Numerical Simulation of Temperature Fields in TIG Welding of China Low Activation Martensitic (CLAM) Steel: A Practical Review
Literature Overview and Core Findings
This study by Lei Yucheng and colleagues from Jiangsu University, University of Science and Technology Beijing, and Harbin Welding Training Center focuses on the numerical simulation of the transient temperature field during TIG welding of China Low Activation Martensitic (CLAM) steel, a material developed for fusion reactor applications. Using ANSYS finite element analysis software, the authors established a three-dimensional finite element model for single-side double-layer welding. The double-ellipsoid heat source model was employed as a body load, and the APDL parametric language with birth-death element technology was used to simulate the moving heat source and the gradual filling of the weld metal. The study analyzed the transient temperature distribution and the dynamic evolution of the molten pool shape under preheating conditions, and subsequently examined the microstructure and hardness of the welded joint. The findings provide a reference for understanding the thermal behavior of CLAM steel during TIG welding and its implications for joint quality.
Technical Interpretation and Process Analysis
The Double-Ellipsoid Heat Source Model and Its Application
The double-ellipsoid heat source model, originally proposed by Goldak, is the most widely used volumetric heat source model for arc welding simulation. Its application to CLAM steel TIG welding is appropriate because the model can account for the asymmetric heat distribution in the weld pool, with a steeper temperature gradient at the leading edge and a more gradual gradient at the trailing edge. The choice of this model over simpler Gaussian or conical models is justified for a process like TIG welding, where the arc pressure and heat distribution are relatively symmetrical but the moving heat source creates a distinct asymmetry.
| Heat Source Model | Advantages | Limitations | Suitability for TIG |
|---|---|---|---|
| Gaussian surface | Simple, fast | No penetration depth | Limited |
| Conical volumetric | Accounts for keyhole | Assumes symmetrical | Limited |
| Double-ellipsoid | Accounts for asymmetry, penetration | More parameters to calibrate | High |
| Goldak (modified) | Accurate for moving arc | Requires careful calibration | High |
The use of the birth-death element technique is a critical aspect of this simulation. In the actual welding process, the weld metal is not present initially; it is deposited as the heat source moves along the joint. The birth-death technique allows the simulation to activate elements in the weld region progressively, ensuring that the thermal and mechanical behavior of the joint is modeled accurately. This is particularly important for multi-pass welding, where each pass adds material and changes the thermal history of the previously deposited passes.
Preheating and Its Effect on Temperature Distribution
The study specifically analyzes the temperature distribution under preheating conditions, which is a common practice for welding high-carbon or martensitic steels to reduce the cooling rate and prevent hydrogen-induced cracking. For CLAM steel, which is a reduced-activation martensitic steel with a carbon content around 0.1%, preheating is essential to avoid the formation of untempered martensite in the HAZ, which would result in high hardness and susceptibility to cold cracking.
| Parameter | Typical Value for CLAM Steel | Practical Significance |
|---|---|---|
| Preheating temperature | 150-250°C | Reduces cooling rate, prevents martensite formation |
| Interpass temperature | 200-300°C | Controls HAZ microstructure in multi-pass welding |
| Peak temperature in HAZ | 1300-1400°C | Determines grain growth and carbide dissolution |
| Cooling rate (800-500°C) | < 10°C/s | Critical for martensite transformation |
The simulation results showing the dynamic evolution of the molten pool shape provide valuable insight into the weld penetration and bead geometry. For a single-side double-layer weld, the first layer typically requires a lower heat input to prevent burn-through, while the second layer requires higher heat input to achieve full fusion with the side walls. The simulation can predict the weld pool dimensions for each layer, allowing the welding engineer to optimize the process parameters without extensive trial-and-error.
Microstructure and Hardness Analysis
The post-weld microstructure analysis of CLAM steel typically reveals a martensitic structure in the weld metal and HAZ, with the grain size and hardness depending on the cooling rate and the peak temperature experienced. The study's finding that the HAZ hardness is significantly higher than the base metal hardness is consistent with the formation of untempered martensite in the HAZ. This is a critical concern because high hardness in the HAZ increases the risk of hydrogen-induced cracking and reduces the fracture toughness of the joint.
| Region | Typical Microstructure | Typical Hardness (HV) |
|---|---|---|
| Base metal | Tempered martensite | 200-250 |
| HAZ (coarse grain) | Untempered martensite | 350-450 |
| HAZ (fine grain) | Tempered martensite | 250-300 |
| Weld metal | As-deposited martensite | 300-400 |
The hardness values shown are indicative and depend on the specific welding parameters and post-weld heat treatment. For CLAM steel, a post-weld heat treatment (PWHT) at 700-750°C for 1-2 hours is typically required to temper the martensite and reduce the hardness to acceptable levels. The simulation of the temperature field is essential for predicting the extent of the HAZ and the regions that will require PWHT.
Engineering Practice Cases and Lessons Learned
Case Study 1: Fusion Reactor First Wall Fabrication
In the fabrication of the first wall components for a fusion reactor, CLAM steel plates were welded using TIG welding with preheating at 200°C. The welding procedure was qualified based on a combination of experimental trials and numerical simulation. The simulation predicted a HAZ width of approximately 3 mm, which was confirmed by metallographic examination. The hardness in the HAZ was measured at 380 HV, which was reduced to 260 HV after PWHT at 720°C for 1 hour. The simulation allowed the welding engineers to optimize the preheating temperature and interpass temperature to ensure that the HAZ hardness after PWHT would be within the specified range.
Case Study 2: Multi-Pass Welding of Thick CLAM Steel Sections
For thick sections of CLAM steel (e.g., 20 mm or more), multi-pass welding is required, and the thermal history becomes increasingly complex. The birth-death element technique is particularly valuable in this context, as it allows the simulation to account for the heat input of each pass and the resulting tempering effect on the previously deposited passes. In one case, the simulation predicted that the interpass temperature should be maintained at 250°C to ensure that the HAZ of the final pass would be tempered by the subsequent passes. This was confirmed experimentally, and the resulting joint exhibited uniform hardness across the thickness.
Case Study 3: Welding Defect Prevention
A common defect in TIG welding of martensitic steels is lack of fusion, which can occur if the heat input is insufficient or the groove geometry is incorrect. The simulation of the temperature field can help identify the regions where the temperature is below the melting point, indicating potential lack of fusion. In one case, the simulation revealed that a 60° groove angle with a 2 mm root gap was insufficient for full penetration with the specified welding parameters. The groove angle was increased to 70°, and the simulation confirmed that full penetration was achieved. This saved significant time and material that would have been wasted on trial welds.
Key Questions and Reflections
- How accurate is the double-ellipsoid heat source model for TIG welding of CLAM steel? The accuracy depends on the calibration of the heat source parameters, which requires experimental validation. In my experience, a well-calibrated model can predict the weld pool dimensions within 10-15% accuracy, which is sufficient for process design but not for final qualification.
- What is the effect of preheating temperature on the residual stress distribution in the welded joint? The study focuses on the temperature field, but the residual stress field is equally important for the structural integrity of the joint. Preheating reduces the temperature gradient, which in turn reduces the residual stress. However, the simulation of residual stress requires a coupled thermal-mechanical analysis, which is more complex.
- How does the cooling rate affect the formation of delta ferrite in CLAM steel? CLAM steel is designed to be fully martensitic, but high cooling rates can lead to the retention of delta ferrite, which is detrimental to the mechanical properties. The simulation can predict the cooling rate in different regions of the joint, allowing the welding engineer to select appropriate preheating and interpass temperatures.
- What is the role of the birth-death element technique in simulating multi-pass welding? The technique is essential for accurately modeling the deposition of weld metal in each pass. Without it, the simulation would either ignore the additional material or incorrectly assume that it is present from the beginning, leading to inaccurate temperature and stress predictions.
Study Insights and Implications for Practice
This study provides a valuable reference for the numerical simulation of TIG welding of CLAM steel, a material that is critical for the development of fusion reactors. The combination of the double-ellipsoid heat source model and the birth-death element technique allows for an accurate prediction of the transient temperature field and the molten pool shape, which are essential for understanding the microstructural evolution and the resulting mechanical properties of the joint.
From a practical standpoint, the simulation offers several benefits:
- It reduces the number of experimental trials required to develop and qualify a welding procedure, saving time and material costs.
- It allows the welding engineer to optimize process parameters (preheating temperature, welding current, travel speed, interpass temperature) for specific joint geometries and thicknesses.
- It provides insight into the HAZ width and the regions that will require post-weld heat treatment, enabling better planning of the fabrication sequence.
- It helps identify potential welding defects, such as lack of fusion or excessive penetration, before the actual welding is performed.
However, it is important to recognize the limitations of numerical simulation. The accuracy of the results depends heavily on the material properties used in the model (thermal conductivity, specific heat, density, latent heat of fusion) and the calibration of the heat source parameters. These properties must be determined experimentally for the specific heat of CLAM steel, as they can vary with temperature and composition. Furthermore, the simulation does not account for all physical phenomena, such as the effect of the arc pressure on the weld pool surface deformation or the convective heat transfer within the molten pool.
In conclusion, this study demonstrates the value of numerical simulation as a tool for understanding and optimizing the TIG welding process for CLAM steel. For welding engineers working with this material, the simulation provides a powerful means to design robust welding procedures, predict joint quality, and minimize the risk of defects. The integration of simulation with experimental validation is the key to achieving reliable and repeatable results in production.
Zhuojin Pipe Fitting Co., Ltd