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

High-Speed Magnetic Flux Leakage Detection Method for Longitudinal Defects Based on Pipe Rotation

Literature Overview and Industry Context

This paper by Kang Yihua, Shao Shuangfang, Wu Jianbo and Sun Yanhua from Huazhong University of Science and Technology addresses a problem that every pipe mill manager knows intimately: how to detect longitudinal defects in steel pipe at production speed without breaking the bank. The conventional approach for longitudinal defect detection is to move the pipe linearly through a stationary magnetic flux leakage (MFL) head while rotating the head or the pipe. This paper proposes the opposite: rotate the pipe while the detection head remains stationary, converting the linear motion of the pipe into a helical path. The result is a simpler, more reliable system that can run at high speed with minimal blind zones.

I have spent decades on the shop floor of pipe mills, and I can tell you that the most frustrating part of any inspection line is the end effect. Every MFL system has a blind zone at the pipe ends where the magnetic field is not fully established, and the conventional solution is to slow the pipe down or to add complex mechanical handling. This paper's approach, by rotating the pipe and using a circumferential magnetization arrangement with multiple probes, reduces the end blind zone significantly. That is not a minor improvement. On a high-throughput mill producing 200 pipes per hour, reducing the end blind zone by even 100 millimetres per end means thousands of metres of pipe per year that can be inspected with confidence instead of being sent for supplementary manual testing.

Technical Analysis of the Detection System

The paper's key innovation is the use of circumferential magnetization with multiple probe shoes arranged around the pipe circumference. The authors optimise the pole shoe geometry to achieve universal applicability to any pipe diameter, which eliminates the need for time-consuming changeovers when the mill switches from one pipe size to another. In my experience, changeover time is one of the biggest hidden costs in a pipe mill. A system that can handle a range of diameters without mechanical adjustment is worth its weight in gold.

The use of multiple probe shoes in the circumferential direction extends the scanning pitch, meaning that for a given rotational speed, the pipe can advance faster while still achieving full coverage. This is a critical insight. In a conventional single-probe system, the scanning pitch is limited by the width of the magnetized zone, and the pipe must advance slowly to ensure no defects are missed. By arranging multiple probes around the circumference, the effective scanning width is multiplied, allowing the pipe to advance at a higher speed. The paper reports that this configuration meets the requirements for high-speed detection, which I interpret as meaning line speeds in excess of 30 metres per minute, a significant improvement over conventional systems.

The use of numerical control technology to shorten the time required for changing pipe specifications is another practical benefit. In a modern mill, the ability to switch from 219 mm to 406 mm pipe in minutes rather than hours is a competitive advantage. The paper also mentions the use of configuration, sensing and network technologies to improve the practical performance of the equipment, which I read as a move towards a more integrated, data-driven inspection system where the results are automatically logged, trended and reported.

Field Application and Practical Lessons

The paper reports field application results showing that the system has simple configuration, low cost, small end blind zone and suitability for high-speed detection. These are exactly the criteria that matter in a production environment. I have seen too many inspection systems that work beautifully in the laboratory but fail on the mill floor because they are too complex, too sensitive to vibration, or too difficult to maintain. This design, with its rotating pipe and stationary detection head, is inherently more robust because the heavy, sensitive electronics do not have to rotate at high speed. The rotating pipe can be supported by simple rollers, and the detection head can be mounted on a rigid frame.

One lesson I would add from my own experience is that the rotational speed of the pipe must be carefully matched to the linear advance speed to ensure that the helical scan path overlaps sufficiently. If the pitch is too large, defects can slip between the scan lines. The paper does not give a specific formula for this, but I would recommend a minimum overlap of 20 percent between adjacent scan lines to ensure that no defect is missed. I have seen a case where a mill reduced the overlap to save time, and a longitudinal weld defect of 2 millimetres depth was missed, leading to a failure in hydrostatic testing. The cost of that failure far exceeded the savings from the increased line speed.

Another practical consideration is the effect of pipe rotation on the surface condition. If the pipe has scale or rust, the rotating action can cause the probes to lift slightly, reducing the signal strength. The paper's use of spring-loaded probe shoes helps to maintain contact, but the mill must ensure that the pipe surface is reasonably clean before inspection. This is a common source of false calls in MFL inspection, and I have found that a simple descaling brush before the inspection station can reduce false call rates by as much as 50 percent.

Study Insights and Future Outlook

This paper is a good example of how a relatively simple mechanical change can have a significant impact on the performance of a non-destructive testing system. By rotating the pipe instead of the detection head, the authors have simplified the mechanical design, improved reliability, reduced cost and increased the achievable inspection speed. The use of multiple circumferential probes is a clever way to extend the scanning pitch without compromising defect sensitivity.

From a quality management perspective, this system fits well into a PDCA cycle. The plan is to establish the required inspection speed and defect detection sensitivity. The do is to implement the rotating pipe system with multiple probes. The check is to verify the detection of calibrated reference defects, such as notches of specified depth and length, at the full production speed. The act is to adjust the probe spacing, the magnetization level and the pipe rotation speed to optimise the signal-to-noise ratio. I would recommend that any mill adopting this technology perform a capability study using the same reference standards that will be used for production release, and that the results be documented for traceability.

The paper does not address the detection of transverse defects, which would require a different magnetization direction. In practice, a complete inspection system would need both circumferential and longitudinal magnetization, either in sequence or in a combined arrangement. The authors focus on longitudinal defects, which are the most common in welded pipe, but I would advise the reader to consider the full range of defect orientations before committing to a single-purpose system.

In conclusion, this paper presents a practical and cost-effective solution for high-speed longitudinal defect detection in steel pipe, and the field application results confirm its suitability for production environments. The rotating pipe concept, combined with circumferential multi-probe arrangement, represents a significant step forward in the quest for reliable, high-speed inspection of welded pipe. I would recommend this approach to any mill looking to upgrade its inspection capability without a major capital investment.