ZHUOJIN-LOGOZhuojin Pipe Fitting Co., Ltd
Zhuojin Pipe Fitting Co., Ltd
STEEL PIPE · FITTING · WELDING  |  NOTES FROM A 30-YEAR INDUSTRY VETERAN

MIG Welding Torch Position Sensing Control System: A Practical Review

Literature Overview and Core Contribution

The paper by Chen Kexuan, Li Heqi, Li Chunxu, and Wang Xijing from Lanzhou University of Technology addresses a persistent headache in automated MIG welding: how to keep the torch centered on the joint when the fit-up is imperfect, the part is distorted, or the fixture is not repeatable. The team built a closed-loop control system using an 80C552 microcontroller as the brain, a servo motor as the actuator, and a rotating arc scanner as the position sensor. The rotating arc scanner is the key innovation here. Instead of relying on a separate optical or tactile sensor that can be fooled by arc light, spatter, or surface scale, the system uses the welding arc itself as the sensing element. By rotating the torch tip in a small circular pattern, the arc length—and therefore the welding current or voltage—varies as the torch moves off the joint centerline. The microcontroller reads these variations, computes the lateral and vertical error, and commands the servo motor to correct the torch position. The reported results show both height and lateral tracking errors under 1 mm, with good real-time response and strong anti-interference performance.

Technical Interpretation and Process Window

From a practical standpoint, the significance of this work lies in the fact that it demonstrates a viable path to seam tracking without adding expensive hardware to the welding cell. The rotating arc sensor is essentially a mechanical modulation of the arc itself, which means it works on any conductive workpiece and is immune to the optical interference that plagues camera-based systems. The control loop operates at a frequency that is high enough to compensate for the slow drift caused by thermal distortion and low-frequency weld pool oscillation, but not so high that it amplifies arc noise. The 1 mm accuracy is adequate for most V-groove and U-groove joints in pipe welding, where the root gap typically ranges from 2 to 4 mm and the acceptable misalignment is around 1.5 mm. In real production, this means the system can hold the torch within the groove even when the pipe is oval, the tack welds are uneven, or the fit-up gap varies by a millimeter or two along the seam.

Parameter Reported Value Practical Acceptance Criterion
Height tracking error < 1 mm ≤ 1.5 mm for root pass in V-groove
Lateral tracking error < 1 mm ≤ 1.5 mm for root pass in V-groove
Control loop response Real-time, no lag noted Must respond within 100 ms for 1 mm step
Anti-interference Strong against arc light, spatter Must reject 50 Hz mains hum and high-frequency inverter noise
Sensor type Rotating arc scanner No external sensor, works on any conductive material

Integration with Manufacturing Practice

In my experience running pipe welding lines, the most common cause of root pass defects is not the welding parameters themselves but the inability of the operator or the automation to follow the joint. When we weld large-diameter LSAW pipe in the longitudinal seam, the edges are prepared by milling and the gap is controlled, but the pipe still has residual ovality and the seam can wander by a few millimeters over a 12-meter length. A rotating arc sensor system like the one described here would allow us to run the root pass at higher speed with confidence, because the torch would automatically stay centered. The same principle applies to circumferential girth welds on line pipe, where the pipe is rotated under a stationary torch. If the pipe is not perfectly round or the rotation axis is not perfectly aligned, the torch-to-work distance changes, and the arc sensor compensates. This is exactly the kind of practical solution that saves hours of grinding and rework on site.

Key Questions and Reflections

One question that the paper does not fully address is the behavior of the rotating arc sensor when the joint has a backing strip or a copper backing bar. The arc length modulation depends on the workpiece surface geometry, and a flat backing strip changes the signal characteristics. In my experience, this is a real limitation for pipe welding with backing rings. Another practical concern is the wear of the rotating mechanism. The scanner involves mechanical rotation at the torch tip, which means the contact tip and the shielding gas nozzle are subjected to continuous motion. In a production environment with high duty cycle, this can lead to premature wear of the contact tip and inconsistent arc behavior. A more robust approach for heavy industrial use would be to use a magnetic or optical sensor for the coarse tracking and the arc sensor for the fine adjustment, but the paper demonstrates that the arc sensor alone can achieve the required accuracy for many applications.

Study Insights and Practical Recommendations

The takeaway for the pipe and fitting industry is that seam tracking does not have to be a complex, expensive add-on. A well-designed arc sensor with a simple microcontroller can deliver sub-millimeter accuracy, which is sufficient for most root pass and fill pass operations. For shop fabrication of pipe fittings, where the joints are often short and the fit-up is reasonably good, this system can be integrated into an existing welding carriage with minimal modification. For field welding of girth seams, the system offers the advantage of being robust against wind, dust, and varying light conditions, which are the enemies of optical sensors. I would recommend that any engineer considering automation of MIG welding for pipe or fitting fabrication evaluate the rotating arc sensor approach, particularly if the parts have moderate fit-up variation and the production volume justifies the investment in the control hardware. The 1 mm accuracy is a solid baseline, and with proper tuning of the control loop gains, it can likely be improved further.

Concluding Remarks

This work provides a practical, cost-effective solution to a problem that has plagued automated welding for decades. The rotating arc sensor eliminates the need for external sensors, works reliably in harsh environments, and achieves tracking accuracy that meets the requirements of most pipe and fitting welding applications. The integration of a microcontroller-based controller with a servo motor actuator demonstrates that even modest hardware can deliver robust performance when the control algorithm is well designed. For the industry veteran, this paper is a reminder that sometimes the simplest solution is the best one, and that the welding arc itself can be a powerful sensing tool if we take the time to understand its behavior. The reported results are a solid foundation for further development, and I would encourage any welding engineer to consider this approach when faced with the challenge of automating a MIG welding process where joint tracking is a bottleneck.