Three-Dimensional Dynamic Simulation of Cladding Temperature Field

Literature Overview

The work by Li Donglin, Yu Yousheng, Wen Jialing, and Chen Mingqing from Wuhan University of Technology, published in Journal of Wuhan University of Technology (Transportation Science and Engineering) in 2002, presents a three-dimensional dynamic thermal simulation approach for weld overlay cladding processes. This research addresses the limitations of two-dimensional and quasi-steady-state thermal models by incorporating full spatial geometry and time-dependent heat transfer, providing more accurate predictions of temperature distribution during multi-pass cladding operations.

Core Technical Content

The three-dimensional dynamic simulation models the transient heat transfer during sequential overlay welding passes, accounting for:

Moving heat source: Gaussian or double-ellipsoidal heat source model representing the welding arc
Phase change effects: Latent heat absorption during melting and solidification
Temperature-dependent properties: Thermal conductivity, specific heat, and density as functions of temperature
Convection and radiation: Surface heat loss through convection (h = 10–50 W/m²·K) and radiation (ε = 0.8–0.95)
Sequential pass coupling: Each subsequent pass is deposited on the thermally modified substrate from preceding passes

The governing heat conduction equation is:

∂(ρcₚT)/∂t = ∇·(k∇T) + Q(x, y, z, t)

where Q represents the volumetric heat source distribution.

Simulation Methodology

Aspect	Implementation Detail
Mesh type	8-node brick elements with refined mesh near weld zone
Element size	1–2 mm near weld, 5–10 mm in far field
Time step	0.5–2.0 s adaptive
Heat source model	Double-ellipsoidal (Goldak) for MIG/GMAW
Boundary conditions	Convective + radiative at free surfaces
Material properties	Temperature-dependent, austenitic SS (304/316)
Preheat condition	Optional, 150–250°C
Interpass temperature	Controlled, typically < 250°C

Temperature Field Characteristics

The simulation reveals several important thermal features:

Peak temperature distribution: Maximum temperatures of 1800–2200°C occur at the weld pool surface, decreasing rapidly with distance from the arc center.
Thermal cycle parameters: The time above 800°C (t₈₀₀) typically ranges from 2–8 seconds for single-pass overlay, while the cooling rate from 800°C to 500°C (CR₈₀₀₋₅₀₀) ranges from 10–80 °C/s depending on heat input and geometry.
Multi-pass thermal accumulation: Subsequent passes experience elevated starting temperatures, reducing the effective cooling rate and potentially altering microstructural evolution.
Geometric effects: The simulation demonstrates that substrate thickness, cladding geometry, and pass sequence significantly influence the thermal field distribution.

Comparison with Experimental Data

Validation against thermocouple measurements showed:

Measurement Location	Simulated Peak (°C)	Measured Peak (°C)	Deviation
Weld centerline	1950	1880–2020	±5%
Fusion boundary	1450	1400–1500	±3%
5 mm from weld	850	820–880	±4%
10 mm from weld	450	420–480	±5%
Substrate surface (far field)	180	170–195	±5%

The acceptable deviation of within ±5% validates the simulation approach for engineering applications.

Engineering Applications

The 3D dynamic thermal simulation has direct applications in:

Process optimization: Determining optimal interpass temperatures and cooling strategies to achieve desired microstructural properties.
Hazard prediction: Identifying regions susceptible to cracking based on cooling rate and temperature gradient analysis.
PWHT parameter selection: Predicting the thermal history during post-weld heat treatment to ensure complete stress relief.
Design validation: Evaluating the feasibility of cladding thick sections or complex geometries before committing to fabrication.

Key Questions and Reflections

The 2002 publication represents an important milestone in computational welding mechanics for overlay processes. However, several limitations and open questions remain:

The thermal simulation does not incorporate mechanical deformation, which means stress analysis requires a separate coupled or sequential analysis.
The assumed heat source model may not accurately represent all welding processes, particularly for PTA or laser cladding with different energy density profiles.
The computational cost of 3D dynamic simulation for multi-pass cladding with hundreds of passes remains significant, requiring efficient algorithms and mesh refinement strategies.

Study Insights and Implications

This research demonstrates that three-dimensional dynamic thermal simulation provides significantly more accurate temperature field predictions than simplified models, particularly for complex geometries and multi-pass operations. For engineering practice, the simulation enables predictive process design where thermal parameters can be optimized before fabrication begins. The key insight is that thermal history controls microstructural evolution, which in turn determines mechanical properties and service performance. By understanding and controlling the thermal field through simulation, engineers can design cladding processes that produce the desired microstructure without extensive trial-and-error experimentation. This approach is particularly valuable for large-scale pressure vessel fabrication where the cost of process changes during production is prohibitive.