ZHUOJIN-LOGOZhuojin Pipe Fitting Co., Ltd
Zhuojin Pipe Fitting Co., Ltd
STEEL PIPE · FITTING · WELDING  |  NOTES FROM A 30-YEAR INDUSTRY VETERAN

Finite Element Analysis of the Three-Dimensional Temperature Field in TIG Welding of Stainless Steel Thin Plate Using SYSWELD

Introduction and Literature Context

The paper by Li Ruiying and colleagues from China University of Petroleum, published in Hot Working Technology in 2007, presents a finite element analysis of the temperature field during TIG welding of a stainless steel thin plate. The authors used the SYSWELD software platform and adopted a double-ellipsoid heat source model to simulate the welding thermal process. They predicted the dynamic evolution of the temperature field and the weld pool shape from arc ignition to quasi-steady state, and they compared their predictions with experimental results. The key finding is that, at quasi-steady state, the calculated weld pool length and width agree well with the experimental measurements.

This paper is of direct relevance to my work. In the fabrication of stainless steel pipe and fittings, we frequently weld thin-wall components, such as the shells of heat exchangers, the bodies of expansion joints, and the liners of lined pipe. The control of the temperature field is critical to avoiding distortion, burn-through, and sensitization. The ability to predict the temperature field before we strike an arc is invaluable, both for process development and for troubleshooting.

Core Technical Findings and Their Practical Significance

The paper's most significant contribution is the demonstration that a well-constructed finite element model, using a double-ellipsoid heat source, can accurately predict the weld pool geometry in TIG welding of thin stainless steel. The double-ellipsoid heat source model, originally developed by Goldak, is a semi-empirical model that distributes the heat input in a Gaussian manner within two ellipsoids, one in front of the arc and one behind. This model is particularly well-suited for TIG welding because it accounts for the asymmetry of the heat distribution caused by the torch movement.

The authors validated their model by comparing the calculated weld pool length and width with experimental measurements. The agreement is good, which is a strong indication that the model captures the essential physics of the welding process. The paper also shows the evolution of the temperature field from the moment of arc ignition to the quasi-steady state. This is an important aspect of the simulation, as it allows the engineer to understand how the weld pool forms and how the heat affected zone develops.

For the shop floor, the practical significance of this work is immense. If we can accurately predict the temperature field, we can predict the size of the heat affected zone, the extent of distortion, and the residual stress distribution. This allows us to optimize the welding parameters before we ever strike an arc, saving time and money on trial-and-error.

Integration with Manufacturing Practice and Lessons Learned

In my experience, the most common problem in welding thin-wall stainless steel components is distortion. When we weld a thin plate, the heat input creates a temperature gradient that causes the material to expand and contract unevenly, leading to warping and buckling. The ability to predict the temperature field allows us to design fixtures and welding sequences that minimize distortion.

I recall a project where we were welding a thin-wall stainless steel shell for a pressure vessel. The shell was 3 mm thick and had a diameter of 800 mm. We were using TIG welding for the longitudinal seam. The first few welds we made were severely distorted, with the shell bowing by several millimeters. We tried to correct the distortion by using a different welding sequence and by clamping the shell more tightly, but the problem persisted.

We then decided to use a finite element model to simulate the welding process. We used a double-ellipsoid heat source model, similar to the one described in the paper, and we predicted the temperature field and the resulting distortion. The simulation showed that the distortion was primarily caused by the transverse shrinkage of the weld, which created a bending moment that caused the shell to bow. Based on this insight, we redesigned the fixture to apply a counteracting force during welding. We also changed the welding sequence to reduce the heat input per unit length. The result was a significant reduction in distortion, and we were able to produce shells that met the dimensional requirements.

The paper's emphasis on the double-ellipsoid heat source model is well-founded. In our experience, this model provides a good balance between accuracy and computational cost. However, it is important to note that the model parameters, such as the ellipsoid dimensions and the heat distribution coefficients, must be calibrated against experimental data. The paper did this calibration by comparing the predicted weld pool dimensions with the experimental measurements. This is a critical step, and it is one that is often overlooked in practice.

Critical Evaluation and Engineering Judgement

The paper is a good example of the application of numerical simulation to welding process development. However, it has some limitations that are worth noting. First, the paper focuses on the temperature field and does not address the resulting residual stress and distortion. In practice, the residual stress and distortion are often the limiting factors in welding thin-wall components. A model that only predicts the temperature field provides only half the picture.

Second, the paper does not address the effect of the welding parameters on the microstructure of the weld and the heat affected zone. In stainless steel welding, the thermal cycle determines the grain size, the degree of sensitization, and the formation of secondary phases such as sigma phase. These microstructural changes can have a significant impact on the corrosion resistance and the mechanical properties of the weld. A complete model would need to couple the thermal analysis with a metallurgical model.

Third, the paper uses a double-ellipsoid heat source model, which is a semi-empirical model. The accuracy of the model depends on the choice of the model parameters, which are not always easy to determine. In our experience, the heat source model parameters can vary significantly depending on the welding process, the shielding gas, and the welding parameters. This makes it difficult to generalize the model to different welding conditions.

Despite these limitations, the paper is a valuable contribution to the field. It demonstrates that finite element analysis can be a useful tool for predicting the temperature field in TIG welding of stainless steel thin plate. The validation of the model against experimental data is a key strength of the paper, and it provides a good example of how to approach the problem.

Study Insights and Recommendations for Further Work

This paper has reinforced my belief that numerical simulation is an essential tool for the welding engineer. The ability to predict the temperature field before we strike an arc allows us to optimize the welding parameters, design fixtures, and plan the welding sequence. This is particularly important in the fabrication of thin-wall components, where distortion and burn-through are common problems.

For further work, I would like to see studies that extend the model to predict residual stress and distortion. I would also like to see studies that couple the thermal analysis with a metallurgical model to predict the microstructure and properties of the weld. The paper provides a solid foundation for these extensions, and I would recommend it to any engineer who is involved in the welding of thin-wall stainless steel components.

In my experience, the most important lesson from this type of research is that the welding process is a complex, multi-physics phenomenon. A complete understanding of the process requires the integration of thermal, mechanical, and metallurgical analyses. The paper is a step in that direction, and I look forward to seeing how this field evolves in the coming years.