Numerical Welding Simulation On Cladded Steel

Welding, the most significant of the fusion processes, was more complicated and evolved more slowly. In the early phases of this evolution, fusion welding was largely employed to repair damaged or worn metal parts. Nonetheless, during the First World War, studies on the suitability of this method as a major means of joining steel plate were started. Welding has become increasingly popular for this purpose primarily to its advantages in design freedom, cost savings, decreased total weight, and improved structural performance.

Welding processes have changed and advanced as new materials have been developed. Welding of hot-roll stainless clad steel material has been shown to be the most suitable connecting procedure employed in industries. Consequently, exploring cladded materials certainly involves a whole welding process effective for joining them, using the adequate metal fillers. However, its welding process causes some issues especially on the residual stress distribution which can be produced in the weld zone and their effect on the structure distortion.

1. Welding-induced distortions

Welding distortion is caused by the expansion and contraction of the weld metal and surrounding base metal throughout the welding process's heating and cooling cycles. The distortion in the welding operation is the result of the welded metal’s contraction and the adjacent parent metal during the heating and cooling cycle of the welding process. Many variables, such as physical and mechanical characteristics that vary when heat is applied, impact metal shrinkage and cause deformation during this heating and cooling cycle.

Longitudinal shrinkage, transverse shrinkage and angular distortion are the main types of strain that reflect the stress that occurs in the metal after welding. However, less common distortions can be observed such as rotational distortion, bending and buckling.?

All these kinds of distortions are related to the shrinkage of the weld metal during cooling. They can be subdivided into:

1. Angular distortion: A non-uniform temperature distribution in the through-thickness direction causes transverse uplift. As an example, consider butt-joints with a V-groove.
2. Transverse shrinkage: shrinkage perpendicular to the weld seam
3. Longitudinal shrinkage: shrinkage in direction of the weld seam
4. Rotational distortion: in-plane angular distortion due to the localized thermal expansion and contractions. Very relevant for overlap joints, for instance.
5. Bending distortion: longitudinal uplift. The same causes as angular distortion.
6. Buckling distortion: caused by compressive stresses inducing instabilities in the plates.

Since welding-induced distortion has a direct impact on the quality of the joined components, it is important to minimize distortion in order to satisfy product requirements. Specific weld preparation, improved welding sequencing, pre-bending, heat sinking, thermal straightening, and thermal tensioning might all be used as mitigation strategies. At this point, numerical simulation of welding-induced distortion is a helpful tool that allows for distortion prediction, which leads to a prior description of distortion growth and allows for well-directed optimization of welding-induced distortion evolution in practice.

2. Numerical welding simulation using the finite element method

Over the last 40 years, the finite element approach has been employed to anticipate welding distortion.?The main advantage of the FEA is that is rather simple and easily understandable physically. However, FEM, where the structure is represented as an assembly of finite elements, gained a wide acceptance only over the last decades because of it is capable of solving the nonlinear problem, the complex geometry and practicable in many analysis software ABAQUS, ANSYS, CASTEM, CodeAster, SYSWELD.

To solve the welding problem by the numerical methods, the welding phenomena have to be representable by mathematical formula. The mechanical behavior of welds, according to Goldak, is sensitive to the string connection between heat transfer, microstructure evolution, and thermal stress analysis.

Welding-induced distortion problems are a coupled thermo-mechanical phenomena; however, and since there is a weak coupling between mechanics and heat flow, heat generation by deformation can be ignored, and therefore the most efficient approach is a sequentially coupled thermal and mechanical analysis.

2.1 Thermal simulation

The modeling of the heat source is a key point in a thermal welding simulation. For thermal simulation Goldak classifies the heat source model into five generations by increasing order of difficulty and accuracy. The second generation is the most employed which are prescribed temperature heat source models. These models treat the weld heat source as a sub-domain in which the temperature or specific enthalpy in known as a function of space and time (x,y,z,t). Even if it does not know the physical parameters of melt, in these models, but still gives a good accuracy of temperature distribution outside the melt. The shape of the heat source most commonly use is double ellipsoidal, cylindrical, conical, linear combination of ellipsoidal, conical or other must be used according to the form observed by experiment.

To take into account the movement of the source, and therefore the dissymmetry of the heat source, Goldak proposes a double ellipsoid volumetric source model. The following equation, is the heat input equation for a double ellipsoid volume, with Z is the welding direction.

with ξ = f or r depending on whether z is positive or negative.

Along with advancements in computer performance and simulation techniques, numerical simulations based on FE approaches have recently become widely employed in structural residual stress assessment. Many researchers have studied how to speed up welding simulations to improve analysis accuracy because they are known to be complicated, often requiring accurate representation of geometric features, multiple weld passes, non-linear temperature dependent material properties, thermal and structural boundary conditions.

2.2 Stress simulation

To model the welding process and anticipate distortions and residual stresses, a sequential thermal-stress analysis is usually utilized as the influence of residual stress on heat flow is minor, it may be ignored, unless the boundary conditions are significantly affected by the thermal changes during the welding process. It consists of two analyses: thermal analysis and mechanical analysis.

First, a transient (i.e. time dependent) thermal analysis is performed to identify the temperature history for each node in the mesh. Secondary, mechanical analysis is performed utilizing temperatures read as specified fields from thermal analysis to estimate stresses and displacements owing to changing thermal effects. These stresses are caused by differential contractions that occur as the weld metal solidifies and cools to ambient temperature. Welding, in reality, introduces a significant amount of heat into the material being welded. As a result, the material experiences non-uniform heat distributions, plastic deformations, and phase changes. These modifications result in distinct residual stress patterns in the weld area and the Heat Affected Zone.

The figure above illustrates the anticipated residual stress distribution for a butt-welded carbon steel plate (z as welding direction). At the weld seam, the longitudinal stresses are tensile, whereas in the base material, they are compressive. Transverse stresses are tensile near the weld seam and decrease towards the plate's edges. Transverse tensions are compressed at the beginning and end of the weld.

Below are three illustrations of the same welding procedure with 3 passes.

Temperature evolution:

Stress evolution:

Displacement evolution:

Conclusion

Finally, the significance of welding process modeling is connected to the influence of induced welding residual stresses and distortions on the structural behavior of components under service load. In the lack of solid knowledge on the amount and distribution of residual stresses, it is widely assumed that residual stresses are as high as the material's yield strength, which might lead to overconservatism in design and, as a result, economic issues. The more precise the projected microstructure and residual stress or strain fields, the better one can evaluate the danger of structural damage, such as the creation of fatigue fractures or the initiation of failure.

References:

• Goldak, J., Chakravarti, A., & Bibby, M. (1984). A new finite element model for welding heat sources. Metallurgical transactions B, 15(2), 299-305.
• Cho, Y. R. (2021). Clad Metals: Fabrication, Properties, and Applications.
• Kou, S. (2003). Welding metallurgy. New Jersey, USA, 431(446), 223-225.
• Schwartz, M. M., & Aircraft, S. (1993). Introduction to brazing and soldering. ASM International, ASM Handbook., 6, 109-113.
• Manual, A. W. (2004). Practice and products for stainless steel welding. Avesta Welding, Swedia.
• Falvo, A., Furgiuele, F. M., & Maletta, C. (2005). Laser welding of a NiTi alloy: Mechanical and shape memory behaviour. Materials Science and Engineering: A, 412(1-2), 235-240.
• Depradeux, L. (2004). Simulation numérique du soudage-acier 316L: validation sur cas tests de complexité croissante (Doctoral dissertation, Lyon, INSA).
• Hibbitt, H. D., & Marcal, P. V. (1973). A numerical, thermo-mechanical model for the welding and subsequent loading of a fabricated structure. Computers & Structures, 3(5), 1145-1174.
• Marcal, P. V. (1974). Weld problems(numerical analysis models). Structural mechanics computer programs: Surveys, assessments, and availability, 191-206.
• Friedman, E. (1978). Analysis of weld puddle distortion and its effect on penetration. Welding journal, 57(6), 161.
• Westby, O. (1968). Temperature Distribution in the Work-piece by Welding. Technical University of Norway.
• Masubuchi, K. (2013). Analysis of welded structures: residual stresses, distortion, and their consequences (Vol. 33). Elsevier.