Finite Element Simulation of Temperature Field in Thick Plate Strip Electrode Submerged Arc Surfacing

Literature Overview

This paper by Wang Zhifeng and colleagues from the Harbin Welding Research Institute of the Chinese Academy of Mechanical Sciences, published in Welding Journal (2009, Vol. 30, No. 1, pp. 89-92), presents a finite element analysis (FEA) of the temperature field during strip electrode submerged arc surfacing (SESAS) on thick plates. The study develops a specialized heat source model based on Goldak theory, implements it through FORTRAN subroutines in MSC.MARC software, and validates the numerical results against experimental thermal cycle measurements.

Core Technical Content

Strip Electrode Submerged Arc Surfacing: Process Characteristics

Strip electrode submerged arc surfacing is a high-deposition-rate process that uses a continuous strip electrode (typically 10-50 mm wide) rather than a wire electrode, submerged under a thick flux blanket. This process is particularly suited for:

• Thick surfacing deposits (>25 mm)
• Large component repair and build-up
• Heavy-wear component fabrication
• Nuclear reactor vessel cladding

The process achieves deposition rates of 5-15 kg/h, significantly exceeding conventional submerged arc welding (0.5-2 kg/h), making it economically attractive for thick surfacing applications where wire electrode processes would require excessive processing time.

Goldak Heat Source Model Adaptation

The standard Goldak ellipsoidal heat source model, originally developed for conventional arc welding, was adapted for strip electrode surfacing to account for the unique geometry of the strip electrode heat input. The key modifications include:

The FORTRAN subroutine implementation in MSC.MARC allowed dynamic tracking of the heat source position as the strip electrode moved along the deposition path, enabling realistic simulation of the transient thermal field evolution during multi-pass surfacing.

Validation and Results

The numerical model was validated against:

1. Weld bead geometry measurements: The simulated bead profile (width, reinforcement height, penetration depth) showed good agreement with experimental observations.
2. Thermal cycle curves: Thermocouple measurements at various distances from the weld centerline were compared with simulated temperature-time histories, demonstrating acceptable correlation.

The validated model enables prediction of:

• Peak temperatures at any point in the substrate during deposition
• Cooling rates at various locations (critical for microstructure prediction)
• Thermal cycle history for multi-pass surfacing sequences
• Thermal distortion and residual stress development (when coupled with mechanical analysis)

Practical Significance of Temperature Field Prediction

For thick plate surfacing applications, accurate temperature field prediction is critical for:

• Microstructure control: Cooling rates determine the type and size of phases formed in the surfacing deposit (martensite vs. bainite vs. pearlite)
• Residual stress management: Thermal gradients drive residual stress development, which affects dimensional stability and fatigue performance
• Cracking prevention: High thermal gradients combined with restrained cooling promote cracking, particularly in high-alloy surfacing deposits
• Process optimization: Understanding thermal behavior allows optimization of travel speed, current, and voltage for desired deposit properties

Engineering Practice Integration

In my experience with heavy industrial surfacing applications, particularly for large components such as press rolls, crusher components, and nuclear vessel cladding, the following practical considerations emerge:

1. Model simplification vs. accuracy trade-off: While the Goldak-based model provides good accuracy for temperature field prediction, it does not account for convection and radiation heat losses from the flux surface, which can be significant in thick surfacing applications. For industrial process optimization, these losses should be incorporated.
2. Multi-pass thermal interaction: The paper focuses on single-pass simulation, but in practice, thick surfacing deposits require multiple passes. The thermal interaction between passes significantly affects the final microstructure and residual stress state. Extension of this model to multi-pass simulation would greatly increase its practical utility.
3. Coupled thermal-mechanical analysis: Temperature field prediction is only the first step in understanding surfacing quality. Coupling with mechanical analysis to predict residual stresses, distortions, and cracking susceptibility provides more complete process understanding.
4. Computational efficiency: The FORTRAN subroutine approach, while effective for 2009-era computing, would benefit from modern parallel computing techniques for real-time process monitoring and optimization.

Key Reflections and Study Insights

This paper represents a methodologically sound approach to numerical simulation of surfacing processes, demonstrating the adaptation of established heat source models to non-conventional welding processes. The validation against experimental data provides confidence in the model's predictive capability.

The choice of MSC.MARC as the simulation platform reflects the computational capabilities available in 2009. While modern software packages offer more user-friendly interfaces and built-in welding process modules, the fundamental approach of custom heat source implementation through user subroutines remains relevant for specialized processes not covered by standard software libraries.

The paper's contribution to the field is primarily methodological—demonstrating that established finite element techniques can be successfully applied to strip electrode surfacing with appropriate model adaptation. This opens the door to predictive process design for thick surfacing applications, reducing the reliance on empirical trial-and-error approaches that are time-consuming and costly for large component fabrication.

The validated temperature field model provides a foundation for further development toward:

• Microstructure prediction through coupled thermal-metallurgical analysis
• Residual stress prediction through coupled thermal-mechanical analysis
• Real-time process monitoring and control based on thermal feedback
• Optimization of multi-pass surfacing sequences for thick deposits

This work exemplifies the power of computational methods in understanding and optimizing complex welding processes, particularly where experimental investigation is impractical due to cost or safety constraints.