CLADDING TECHNOLOGY SHANXI CO., LTDSignal Processing Research Based on Rotary Arc TIG Welding
Literature Overview and Research Context
This 2014 publication by Jia Jianping, Liu Yunlong, Chen Jianping, Peng Liang, and Liu Dan from the School of Mechanical and Electrical Engineering, Nanchang University, addresses an innovative approach to TIG welding process monitoring through signal processing techniques applied to rotary arc welding. Supported by the National 863 Program, this research represents a significant advancement in intelligent welding process control. The concept of rotary arc welding—where the torch is rotated during the welding process—introduces unique signal characteristics that differ substantially from conventional stationary TIG welding.
Core Technical Content on Signal Characteristics
Rotary arc TIG welding involves rotating the welding torch around its longitudinal axis while maintaining the arc. This rotation creates periodic variations in the arc geometry, plasma flow, and molten pool dynamics that manifest as characteristic patterns in the welding signals. The key signals analyzed include:
| Signal Type | Frequency Range | Information Content | Sensitivity to Defects |
|---|---|---|---|
| Arc voltage | 0-500 Hz | Arc length, arc stability, gas shielding | High for porosity, low for lack of fusion |
| Arc current | 0-500 Hz | Arc cross-section, melting rate | Moderate for penetration defects |
| Arc sound | 20-20000 Hz | Arc stability, spatter, gas flow | High for shielding gas defects |
| Optical radiation | Broad spectrum | Arc temperature, arc position | High for torch misalignment |
The signal processing methodology employed in this research involves several key steps:
- Signal acquisition: High-frequency sampling of arc voltage and current signals using data acquisition systems with sampling rates of 100-500 kHz.
- Pre-processing: Removal of noise through filtering techniques, including wavelet denoising and adaptive filtering methods.
- Feature extraction: Identification of characteristic parameters such as mean value, standard deviation, peak-to-peak amplitude, and frequency spectrum components.
- Pattern recognition: Application of classification algorithms to distinguish between normal and abnormal welding conditions.
Engineering Practice Applications
The practical significance of this research extends to several important welding applications:
- Weld pool monitoring: The periodic signal variations from rotary arc welding provide real-time information about weld pool geometry and penetration, enabling automatic travel speed adjustment.
- Defect detection: Characteristic signal patterns associated with porosity, lack of fusion, and undercut can be identified in real time, allowing immediate corrective action.
- Process parameter optimization: Signal analysis enables systematic optimization of welding parameters for different material combinations and joint configurations.
- Quality assurance: Signal-based monitoring provides a non-destructive means of ensuring weld quality without reliance on post-weld inspection alone.
For cladding and overlay welding applications, the rotary arc technique offers particular advantages in achieving uniform overlay thickness and consistent metallurgical properties across the cladded surface. The signal processing approach enables closed-loop control of the overlay process, which is critical for maintaining the required dilution ratio in dissimilar metal cladding operations.
Study Insights and Methodological Reflections
This research represents a paradigm shift in welding process control—from reactive quality inspection to proactive process monitoring and control. The signal processing approach described here can be integrated with modern control systems to create adaptive welding processes that automatically compensate for variations in material properties, joint fit-up, and environmental conditions.
The methodology also has important implications for welder training and qualification. By establishing clear signal signatures associated with good and poor weld quality, objective criteria for welder performance can be defined, complementing the traditional visual and destructive testing approaches. This represents a significant advancement in welding quality assurance that aligns with the industry's increasing demand for traceable, data-driven manufacturing processes.