Dynamic Simulation of Dissimilar Material Overlay Welding Temperature Field Using ANSYS

Literature Overview

This paper by Dong Xiaoqiang and Duan Hongyan from the School of Materials Science and Engineering, Shenyang University of Technology, published in the Journal of Shenyang University of Technology (2008, Vol. 30, No. 5, pp. 551-554), presents a numerical simulation methodology for predicting the transient temperature field during overlay welding of dissimilar materials. The study addresses a significant gap in welding simulation literature, where most temperature field analyses are limited to same-material welding scenarios. The authors employ ANSYS finite element analysis software with APDL programming to model the moving heat source and dynamically simulate the temperature distribution during overlay welding of dissimilar materials.

Technical Significance of Dissimilar Material Welding Simulation

Overlay welding inherently involves the joining of dissimilar materials: the base metal and the weld metal have different thermal properties (conductivity, specific heat, density, thermal expansion coefficient). This difference creates complex thermal gradients at the interface that cannot be accurately captured by same-material simulation models. The consequences of inaccurate temperature field prediction include:

Consequence	Impact on Weld Quality
Incorrect peak temperature prediction	Misestimation of grain growth and phase transformations
Inaccurate cooling rate calculation	Wrong prediction of microstructure (martensite, bainite, pearlite)
Poor thermal stress estimation	Underprediction of cracking risk
Incorrect dilution rate calculation	Wrong composition prediction of weld metal
Inadequate HAZ characterization	Missed identification of soft zones or hard zones

The ability to accurately simulate the temperature field during dissimilar material overlay welding is therefore essential for predicting weld microstructure, mechanical properties, and service performance.

Simulation Methodology

The authors developed a simulation approach incorporating several advanced finite element techniques:

Three-dimensional mesh generation: The workpiece geometry is discretized using a 3D finite element mesh. The choice of element type (tetrahedral, hexahedral, or brick elements) and mesh density directly affects computational accuracy and efficiency.
Mesh adaptive technology: The weld metal region is automatically refined and generated as the welding process progresses. This adaptive meshing technique ensures adequate resolution in the rapidly changing temperature field region while maintaining computational efficiency in regions with minimal thermal gradients.
Moving heat source model: The welding heat source is modeled as a moving heat flux applied to the mesh surface. The heat source model may be Gaussian, double-ellipsoidal (Goldak), or a custom formulation. The heat source parameters (power, radius, trailing/falling factors) are calibrated against experimental thermocouple measurements.
APDL programming: The ANSYS Parametric Design Language (APDL) is used to automate the simulation process, including heat source movement, mesh adaptation, and boundary condition application. This scripting capability enables efficient batch processing of multiple welding parameter sets.
Material property modeling: The thermal properties of both the base metal and the weld metal are defined as functions of temperature, accounting for phase transformations (e.g., melting, solidification, and solid-state transformations). The dissimilar material interface is handled by assigning different material properties to different element sets.

Validation and Results

The simulation results were validated against experimental temperature measurements obtained using thermocouples embedded at strategic locations on the workpiece. The close agreement between calculated and measured temperatures confirms the accuracy of the simulation methodology. Key findings from the simulation include:

The temperature distribution is asymmetric due to the different thermal conductivities of the base and overlay materials.
The cooling rate at the weld metal-base metal interface is significantly influenced by the thermal conductivity ratio of the two materials.
The peak temperature in the weld pool varies with the relative position of the heat source and the interface.
The thermal cycle experienced by the heat-affected zone (HAZ) is more severe when the base metal has lower thermal conductivity (e.g., stainless steel on carbon steel).

Process Optimization Implications

The temperature field simulation provides a powerful tool for optimizing overlay welding parameters:

Parameter	Optimization Target	Simulation-Based Approach
Welding current	Control peak temperature and cooling rate	Vary heat input, monitor peak T and cooling rate
Travel speed	Minimize dilution while ensuring fusion	Balance heat input with deposition rate
Heat source geometry	Control weld pool shape and penetration	Adjust Goldak parameters for desired profile
Preheat temperature	Reduce thermal gradient and cracking risk	Simulate thermal stress with and without preheat
Multi-pass strategy	Control interpass temperature and residual stress	Simulate sequential pass deposition with cooling

Engineering Practice Integration

In my engineering practice, the integration of numerical simulation with overlay welding process development follows a systematic workflow:

Define the problem: Identify the dissimilar material combination, service conditions, and performance requirements.
Develop the simulation model: Create a geometric model, assign material properties, define boundary conditions, and calibrate the heat source model.
Validate the model: Compare simulation results with experimental temperature measurements and adjust model parameters until acceptable agreement is achieved.
Optimize parameters: Use the validated model to explore the parameter space and identify optimal welding conditions.
Experimental verification: Conduct welding trials using the optimized parameters and verify the predicted properties (microstructure, hardness, dilution, residual stress).
Procedure qualification: Develop and qualify the welding procedure specification (WPS) based on the simulation-optimized parameters.

This approach reduces the number of experimental trials required, accelerates process development, and provides a deeper understanding of the welding metallurgy involved.

Key Questions and Reflections

The study raises several important considerations for practitioners:

How does the simulation account for the effect of the welding flux or shielding gas on the heat input and heat loss? In practice, the flux or gas significantly affects the thermal efficiency of the welding process.
What is the computational cost of the mesh adaptive technique? For large-scale overlay welding applications (e.g., full duct overlay), the computational resources required may be substantial.
How does the simulation handle the phase transformation during solidification and cooling? The latent heat of fusion and solid-state transformations significantly affect the temperature field and must be accurately modeled.
Can the simulation be extended to predict residual stress and distortion? Temperature field data is a prerequisite for mechanical analysis, and the integration of thermal and mechanical simulations provides a more complete picture of weld quality.

Study Insights and Implications

The research by Dong and Duan demonstrates the value of numerical simulation in addressing the unique challenges of dissimilar material overlay welding. The development of a validated simulation methodology provides welding engineers with a powerful tool for process optimization and quality prediction. The use of mesh adaptive technology and APDL programming represents a practical approach to balancing computational accuracy with efficiency. For contemporary engineering practice, this methodology can be extended to include coupled thermal-mechanical analysis, microstructure prediction models, and multi-physics simulations that incorporate fluid flow in the weld pool. The key insight is that simulation should not replace experimental validation but should guide and inform experimental design, reducing the cost and time of process development while improving the understanding of the underlying metallurgical phenomena. As computational resources continue to improve, the integration of simulation into routine overlay welding process development will become increasingly standard practice.