Elsevier

NDT & E International

Volume 43, Issue 4, June 2010, Pages 323-328
NDT & E International

Detection of back-side pit on a ferrous plate by magnetic flux leakage method with analyzing magnetic field vector

https://doi.org/10.1016/j.ndteint.2010.01.004Get rights and content

Abstract

We have developed a magnetic flux leakage (MFL) system using magnetic resistive (MR) sensors for detecting two dimensional magnetic field components, and an induction coil that generates low magnetic field strengths and extremely low frequencies. The signal at each scanned measurement point (i) was divided by the signal strength Mmes,i and phase αi by a lock-in amplifier. Using the strength Mmes,i and phase αi, we calculated the imaginary part of the signal using the common phase β. By optimization of the common phase β to the imaginary part, the analyzed scanning data curve was shown to be effective in estimating the size (depth and diameter) of back-side pits on a ferrous plate. Comparing the two dimensional magnetic field components of leakage, the imaginary part of the y-component parallel to the induced magnetic field was found to be suitable for detecting the back-side pits.

Introduction

Magnetic flux leakage (MFL) testing is one of the most commonly used methods for non-destructive testing, with a long history of applications. To evaluate and inspect flaws in metallic materials using electromagnetic methods, various measurement methods, such as eddy current testing (ECT) [1], MFL [2], [3], [4], [5], [6], [7], [8], remote field testing (RFT) [9], magnetic particle inspection (MPI) and pulsed eddy current (PEC) [10], have been developed. These techniques are categorized into the two methods of eddy current and magnetic flux leakage measurements. ECT, RFT and PEC are based on detecting the eddy current generated by an applied magnetic field. On the other hand, MFL and MPI are based on magnetic flux leakage of the applied magnetic field. Eddy current methods are commonly used to detect flaws at the surface or subsurface of conductive materials, because the eddy current strength decreases with depth due to the skin effect. Recently, low frequency eddy current testing using magnetic sensors, such as a magneto-resistive sensor (MR) [11], [12], [13], [14], [15], [16], [17], Hall sensor [18] or Superconducting Interference Device (SQUID) [19], [20], [21], which can measure low frequency magnetic signals instead of a search coil, have been developed to detect deeper flaws. The skin depth for eddy current generation decreases with increasing frequency, conductivity and permeability. Therefore, the eddy current method is usually applied to non-ferromagnetic materials, since the skin depth in such materials is larger than that in ferromagnetic materials. On the other hand, the magnetic flux leakage method is usually used for ferromagnetic material structures such as steel pipelines, storage tanks, bridges, etc. If a corrosion pit or crack is present in a steel pipe, magnetic flux leakage occurs outside the pipe wall when an applied magnetic field penetrates the wall in a direction parallel to the wall surface. In general, MFL uses a strong magnetization system to yield a measurable magnetic flux leakage from the specimen. The applied magnetic field strength is usually the saturation region of the BH curve, so that a reduction in the material wall thickness will cause a large flux to leak. However, it is difficult to detect anomaly thick-wall, since it requires a stronger magnetization system to attain saturation. One way to resolve this issue is measurement using a highly sensitive magnetic sensor. When low magnetization is applied and a low magnetic leakage signal is detected, optimization of the measurement and analysis method is necessary, because the signal-to-noise ratio is typically worse in this case. Changes in the strength and phase of the magnetic field caused by an anomaly are usually measured in common MFL, and these parameters have been discussed independently. In this study, we developed an MFL system using MR sensors, measured two-dimensional magnetic field components parallel to the sample surface and analyzed the combination of the magnetic field strength and phase.

Section snippets

Experimental

An MFL system, consisting mainly of a sensor probe, a lock-in amplifier, sensor circuits, a current source for an induction coil and a personal computer, was developed (Fig. 1). The sensor probe consists of a half-shaped yoke, induction coils at both ends of the yoke and two-dimensional MR sensors for detecting x- and y-components. The x-axis was directly perpendicular to the line of both yoke ends, and the y-axis was parallel to both yoke ends. The z-axis was perpendicular to the sample

Results and discussion

Magnetic flux leakage was measured from the back-side of the pit. Scanning was carried out with 10 mm steps. Two (x and y) independent magnetic field components were measured, and the curves of each directly measured magnetic field strength Mmes,i, phase αi, and Mim,i derived from curves were compared. Fig. 3 shows Mmes,i, phase αi curves obtained by measurements on the steel plate with pits 30 mm in diameter and 6 mm in depth. From the directly obtained magnetic field strength and phase, some

Conclusions

We compared two dimensional magnetic field components of leakage. The y-component parallel to the induced magnetic field was suitable for detecting the back-side pit of the ferrous plate. Analysis using the imaginary part of the signal was effective for estimating the size of the pit. The peak height and width of the Mim curve showed a proportional relationship to the pit depth and diameter. The developed MFL system does not require a high power current source because saturation of

Cited by (39)

  • Experiment and simulation on testing steel plate with corrosion defects via magnetic flux leakage method

    2022, Journal of Magnetism and Magnetic Materials
    Citation Excerpt :

    They analyzed the signal characteristics related to the degree of damage. The above experimental studies of MFL testing mentioned are around steel corrosion [29–31]. Considering the corrosion that occurs in the practical structure, the size of the zone is directly related to the contact zone between the practical structure and the corrosive medium.

  • Skin depth and detection ability of magneto-optical imaging for weld defects in alternating magnetic field

    2020, Journal of Manufacturing Systems
    Citation Excerpt :

    Some scholars have studied the skin effect of eddy current testing in detail and discussed the feasibility of accurate detection of inner and surface defects at certain excitation frequency [11,12]. A deeper research on MFL under the equivalent of the skin effect caused by MFL and eddy current testing has been carried out [13]. Some scholars also approximate the skin effect of MFL to eddy current testing when using low-frequency excitation to detect internal defects of workpieces [14].

  • Nondestructive magnetic inspection of spot welding

    2018, NDT and E International
    Citation Excerpt :

    Compared with the ultrasonic and radiographic methods, the magnetic method does not provide good morphology information, however it is convenient and low-cost. A possible way to perform magnetic-based testing is the magnetic flux leakage test (MFL), which uses the magnetic flux leakage from the specimen surface when a low frequency magnetic flux is induced [15–17]. A nondestructive spot weld inspection method using MFL with a magnetoresistive sensor was reported recently [18].

View all citing articles on Scopus
View full text