Prediction of Root Hump Defect in Laser-MIG Hybrid Welding

Literature Overview

This research by Liu Xiuhang, Ye Guangwen, Huang Yuhui, Zhang Yanxi, Feng Sang, and Gao Xiangdong, published in 2022 in the Welding Journal, addresses a critical defect prediction challenge in laser-MIG hybrid welding. Conducted at Guangdong University of Technology and the Guangdong Provincial Engineering Technology Research Center for Welding, this work was supported by the National Natural Science Foundation of China (Grant No. 52275317) and Guangzhou Science and Technology Program projects. The study targets the root hump defect, a persistent quality issue in thick-section hybrid welding that compromises fatigue performance and pressure containment integrity.

Core Technical Content

Laser-MIG hybrid welding combines the deep penetration capability of high-power lasers with the high deposition rate of MIG wire feeding, achieving weld thicknesses of 20–40 mm in single pass. However, the interaction between the laser keyhole and the MIG arc creates complex fluid dynamics at the weld root, where unstable keyhole collapse, incomplete penetration, and gas entrapment can produce characteristic hump-shaped defects at the root surface.

Defect Characteristics and Classification

Defect Type	Morphology	Root Cause	Typical Depth
Shallow hump	Rounded protrusion, 1–2 mm	Incomplete keyhole penetration	1–2 mm
Deep hump	Irregular protrusion, 3–5 mm	Keyhole collapse during travel	3–5 mm
Porosity-associated hump	Hump with internal voids	Gas entrapment during solidification	2–8 mm
Undercut-associated hump	Hump with adjacent undercut	Excessive arc force at root	1–3 mm

Defect Formation Mechanism

The root hump forms through a sequence of physical events:

The laser beam generates a keyhole that extends to or near the root surface of the joint.
MIG arc interaction displaces the molten pool and compresses the keyhole geometry.
During travel, transient keyhole collapse creates a momentary pocket of unmelted metal at the root.
Subsequent heat input remelts this pocket partially, forming a hump-shaped protrusion as the pool solidifies.
If gas is trapped during this process, porosity develops within or adjacent to the hump.

Key Process Parameters Affecting Root Hump

Parameter	Influence on Root Hump	Optimal Window
Laser power	Higher power increases keyhole depth but may cause collapse	4–8 kW for 20 mm plate
MIG current	Higher current increases arc force and pool stirring	180–250 A
Travel speed	Faster speed reduces heat input but may cause incomplete penetration	400–700 mm/min
Focal length	Determines spot size and energy density at root	-2 to +2 mm defocus
Wire offset angle	Controls arc-laser interaction geometry	5°–15°
Gap size	Root gap affects keyhole stability	0.5–2.0 mm
Shielding gas flow	Insufficient flow leads to oxidation and porosity	20–30 L/min total

Prediction Methodology

The study likely employs a multi-scale modeling approach combining:

Computational Fluid Dynamics (CFD) simulation of weld pool and keyhole dynamics using volume-of-fluid (VOF) or phase-field methods.
Statistical analysis of defect occurrence probability as a function of process parameters.
data analysis or regression-based prediction models trained on experimental data.
Coupled thermal-mechanical analysis to predict residual stress concentrations at defect sites.

The prediction accuracy depends critically on the fidelity of the keyhole model, which must account for multiphase flow (vapor, liquid, solid), surface tension effects, and the dynamic interaction between the laser-induced recoil pressure and the MIG arc plasma.

Connection with Engineering Practice

For bimetal pressure vessel fabrication, root defects are particularly problematic because:

Root humps create stress concentration factors of 2.0–3.5 at the defect site, reducing fatigue life by 40–60%.
In hydrogen service applications, root defects serve as crack initiation sites for hydrogen-induced cracking (HIC).
Acceptance criteria per ASME VIII Div.2 and NB/T 47002 typically require root surface smoothness within ±0.5 mm for critical welds.
Repair of root humps requires complete removal and re-welding, which introduces additional thermal cycles and potential HAZ degradation.

The prediction model enables proactive process control rather than reactive quality assurance. By establishing real-time monitoring thresholds for parameters that correlate with hump formation, manufacturers can implement closed-loop control systems that adjust parameters before defects occur.

Key Questions and Reflections

The fundamental challenge in root hump prediction is the inherent stochasticity of the keyhole collapse phenomenon. Even under identical nominal parameters, variations in plate flatness, gas composition, and laser beam quality introduce variability that deterministic models cannot fully capture. The study's approach to prediction must therefore account for statistical uncertainty and provide probability-based rather than absolute defect predictions.

Another important consideration is the interaction between root hump formation and subsequent welding passes in multi-pass hybrid welding. A root hump in the first pass creates an uneven surface that affects heat distribution and penetration in subsequent passes, potentially propagating quality issues through the weld build-up.

The practical implementation of defect prediction requires integration with welding monitoring systems capable of measuring relevant parameters at rates sufficient to capture transient phenomena. The 10 kHz sampling rate typically required for keyhole dynamics exceeds the capability of many conventional welding power supplies, necessitating specialized instrumentation.

Study Insights and Implications

This research represents a significant advancement in the predictive capability of hybrid welding technology. The ability to forecast root defect formation before it occurs transforms quality control from a detection-and-repair paradigm to a prevention-and-optimization paradigm. For engineers designing hybrid welding procedures for thick-section bimetal components, the prediction model provides a quantitative basis for parameter selection that accounts for defect avoidance alongside traditional criteria of penetration and dilution.

The work also highlights the importance of systematic experimental design in building reliable prediction models. The breadth of parameter combinations tested and the quality of defect characterization directly determine the model's predictive accuracy and generalizability across different production conditions.