Serial MTJ-Based TMR Sensors in Bridge Configuration for Detection of Fractured Steel Bar in Magnetic Flux Leakage Testing

Thanks to high sensitivity, excellent scalability, and low power consumption, magnetic tunnel junction (MTJ)-based tunnel magnetoresistance (TMR) sensors have been widely implemented in various industrial fields. In nondestructive magnetic flux leakage testing, the magnetic sensor plays a significant role in the detection results. As highly sensitive sensors, integrated MTJs can suppress frequency-dependent noise and thereby decrease detectivity; therefore, serial MTJ-based sensors allow for the design of high-performance sensors to measure variations in magnetic fields. In the present work, we fabricated serial MTJ-based TMR sensors and connected them to a full Wheatstone bridge circuit. Because noise power can be suppressed by using bridge configuration, the TMR sensor with Wheatstone bridge configuration showed low noise spectral density (0.19 μV/Hz0.5) and excellent detectivity (5.29 × 10−8 Oe/Hz0.5) at a frequency of 1 Hz. Furthermore, in magnetic flux leakage testing, compared with one TMR sensor, the Wheatstone bridge TMR sensors provided a higher signal-to-noise ratio for inspection of a steel bar. The one TMR sensor system could provide a high defect signal due to its high sensitivity at low lift-off (4 cm). However, as a result of its excellent detectivity, the full Wheatstone bridge-based TMR sensor detected the defect even at high lift-off (20 cm). This suggests that the developed TMR sensor provides excellent detectivity, detecting weak field changes in magnetic flux leakage testing.

Magnetic charge model of defect in magnetic flux leakage testing

The defect in structures is the major risk to the structural integrity, thus to perform the defect detections and evaluations efficiently is critical in assuring the structural safety. Magnetic flux leakage testing (MFLT) is an important non-destructive testing (NDT) method. Due to its high testing sensitivity and simple operating procedure, it has been widely used in detecting surface and near-surface defects in ferromagnetic components. To improve the accuracy of defect detection, it is necessary to find a suitable source magnetization distribution around a defect, and furthermore, to correlate the defect with the magnetic leakage signals. In this study, a magnetic charge model is proposed, in which both volume- and surface- densities of magnetic charges around a defect are considered. Then, this model is used for the calculation of the magnetic leakage signals caused by a known complex V-shape defect for the verification purpose. The results from the simulation match very well with that from the experiment. It indicates potentials that the magnetic charge model and the associated approach can be applied in MFLT with improved accuracy.