2020 Volume 60 Issue 5 Pages 948-953
Acoustically stimulated electromagnetic (ASEM) waves in thin steel sheets have been investigated for flaw detection. In the ASEM wave technique, magnetization is temporally modulated at the radio frequency (rf) of the irradiated ultrasonic waves through magnetomechanical coupling. The induced rf magnetic fields are detected by a resonant coil antenna and spatial images are obtained by scanning the ultrasound focal spot. In this work, we detected artificial defects (through holes) in thin steel sheets. Specific patterns of magnetic flux density caused by the hole were observed. By improving the sensitivity with a small coil antenna, we visualized a through hole 0.1 mm in diameter with a lift-off of 10 mm.
Applications of the acoustically stimulated electromagnetic (ASEM) wave technique for nondestructive evaluation have been developed. In the ASEM wave technique, an object is irradiated with ultrasound waves and the electromagnetic waves induced by magnetomechanical coupling are detected using a magnetic sensor.1,2,3) Biomedical applications have been investigated and industrial applications are also promising.4,5) In biological tissues such as bone, the electric field modulation due to piezoelectricity generates the ASEM waves. In ferromagnetic metals such as steel, the magnetic field modulation due to piezomagnetism (or magnetostriction)6) is an important factor in the ASEM wave generation. In this paper, we present the experimental results of flaw detection in thin steel sheets using the ASEM wave technique.
Magnetic flux leakage testing (MFLT) is a conventional flaw detection method. A schematic of MFLT is shown in Fig. 1. In MFLT, the magnetic flux leaking from the surface of the steel plate into the air is detected by a magnetic sensor at an applied magnetic field, H, close to the saturation field. The steel plate is usually magnetized by applying a static or low-frequency magnetic field. These conditions lead for the system to be large.7,8) Furthermore the leakage flux is highly attenuated by the distance from the surface of steel and does not propagate as far as electromagnetic waves. Therefore, the distance between the magnetic sensor and the steel surface (lift-off) should typically be a maximum of several millimeters. Maintaining a constant lift-off is often difficult in the measurements during inspections.
Schematic of MFLT.
Electromagnetic waves reach much further than the static leakage field. The restriction of the lift-off could be relaxed if the magnetic properties of steel were transmitted as electromagnetic waves by irradiating it with ultrasound waves. Another advantage of the ASEM wave technique is that the spatial resolution is determined by the size of the ultrasound focal spot, not by the size of the sensor or the lift-off.
The magnetic flux distribution is strongly affected by the presence of flaws or defects because of the large differences in permeability. In addition, the effect of defects is expected to be magnified via the magnetic flux distribution. In the ASEM wave technique, acoustic pressure is locally applied by ultrasound irradiation and the induced radio-frequency (rf) magnetization is detected. Thus, a small defect will be detected even with a large lift-off.
There are three main advantages of the ASEM wave technique for steel flaw detection.
(1) Practically feasible lift-off conditions.
(2) Distribution of intrinsic magnetic flux density, B, inside the steel is imaged rather than the distribution of leakage flux in the air. Therefore, a saturation field does not need to be applied for inspection. Low-field measurements with a compact electromagnet are possible.
(3) Geometric and magnetic anomalies can be examined simultaneously by combining ASEM with ultrasonic testing (UT).
Based on these advantages, we have previously demonstrated the detection of an artificial defect (a through hole 1 mm in diameter) using the ASEM wave technique.4) In addition, the detection of a smaller defect (0.3 mm in diameter) has been also reported.9) In this paper, we improve the signal-to-noise ratio (SNR) using a small coil antenna and detect a through hole 0.1 mm in diameter.
Figure 2 shows a schematic of the experimental setup for measuring ASEM waves. A UT probe and a magnetic sensor (a coil antenna) are placed on the surface of the steel plate and an electromagnet for magnetizing the steel plate is placed on the opposite side. Figure 3 shows schematics of ASEM wave generation. The principle of the ASEM wave technique is as follows:
(i) Acoustic stress is applied locally by ultrasound waves.
(ii) The stress modifies the local magnetic properties of the steel (i.e., B − H curve) through magneto-mechanical coupling, yielding the acoustically induced rf magnetization (ΔB).
(iii) The magnetization generates the rf magnetic field in the surrounding environment.
(iv) The emitted rf magnetic field is picked up by a resonant coil antenna tuned to the ultrasound frequency.
Schematic of the ASEM wave technique. (Online version in color.)
Principle of ASEM wave generation. (Online version in color.)
A coil is used because its sensitivity for high-frequency magnetic fields is greater than that of conventional magnetic sensors. The magnetic properties of steel are changed by strain,6) and the relationship between the ASEM response signal and the B − H characteristics have been reported previously.4,5) The ASEM waves are observed even if there are no flaws or defects. The magnetic field dependence of the ASEM signal intensity is shown in Fig. 4. When the magnetization is increased at low fields (high-permeability region), the domains are easily rearranged when stress is applied. However, above a critical field, the domain wall movement is complete, and the stress sensitivity of the magnetization decreases as the magnetic field increases in the saturated area.
ASEM signal intensity versus magnetic field. (Online version in color.)
Next, we describe flaw detection based on numerical calculations. When there is an artificial defect (a through hole) in the steel plate, the magnetic flux distribution is different from that in the defect-free part of the plate. Figures 5 and 6 show the numerically calculated two-dimensional (2D) images of magnetic flux density, B, under an applied magnetic field, H. In Fig. 5 (without a hole), we calculated B in a 2.5 × 2.5 mm area (1/4 area). In Fig. 6 (with a hole), we extended it to a 5 × 5 mm area using symmetry. Figures 6(a) and 6(b) show the calculated results in a saturation field and a lower field (70% of the saturation field), respectively. The flux density is uniform in the defect-free part of the plate (Fig. 5), whereas specific spatial patterns in the magnetic flux density due to the defect are clearly observed (Fig. 6(b)). The specific pattern is observed even in the low field (Fig. 6(a)), which indicates that it is not necessary to apply a high saturation field in the ASEM wave technique. Therefore, a compact electromagnet is suitable for the ASEM wave technique.
Numerically calculated map of magnetic flux density in a steel plate without a hole. (Online version in color.)
Numerically calculated map of magnetic flux density in a steel plate with a hole 0.5 mm in diameter for (a) a saturation field and (b) a low field (70% of the saturation field).
Steel sheets for cans 0.16 mm thick (size: 80 × 80 mm) were prepared by introducing artificial defects (through holes, diameters of 0.5, 0.3, and 0.1 mm) by electric discharge machining.
A 10 MHz burst wave (three waves) is transmitted from a pulser-receiver to a focus UT probe with a polystyrene delay line (10 MHz, focal length: 24 mm, focal diameter: about 1 mm). Two coil antennas (Coils A and B) consisting of 0.5 mm copper wire are used. Coil A (diameter: 15 mm, single layer of 20 turns) is a standard coil, as used in Refs. 4 and 5). Coil B (diameter: 5 mm, triple layer of 13, 12, and 11 turns) is a smaller coil to improve the sensitivity. The central axis of the coil is tilted by an angle of 10°–15° with respect to the surface of the steel plate (Fig. 2). The distance between the coil and ultrasound focal spot on the steel surface is about 10 mm, which corresponds to the lift-off. The block diagram of the measurement system is shown in Fig. 7. The antennas are tuned to about 10 MHz using a resonant circuit. The signal is amplified by 86 dB and averaged over 1000 pulses in an analogue-to-digital converter. The ultrasound focal point is scanned by moving the sample with an XY stage, where a gap of 4 mm between the magnetic poles of the electromagnet (850 turns) and the steel plate is created.
Block diagram of the ASEM measurements. BPF: band-pass filter, A/D: analogue-to-digital converter.
Typical real-time waveforms of the UT (echo) and ASEM signal are shown in Fig. 8. We observe a signal at the midpoint t = τecho/2 = 12 μs between the excitation and the echo pulses, where τecho is the delay time of the echo signal reflected on a steel plate. This is identified as the ASEM waves emitted from the steel plate.4,5) The signal intensity of the ASEM waveform is defined as the integral of the absolute value for the signal voltage, and the integrated time, Δτ, is shown in Fig. 9. We measured the ASEM waveforms using Coils A (Fig. 10(a)) and B (Fig. 10(b)). The SNR is considerably higher in Coil B. Because Coil A has a large detection area, unintended signals are detected; thus, the SNR is lower due to the noise from the UT probe or the servo-control noise from the XY stage.
Real-time waveforms of the (a) UT and (b) ASEM signals.
Enlarged ASEM waveform.
ASEM waveform measured by (a) Coil A and (b) Coil B. Arrows indicate the ASEM signal from the steel plate.
The increase of the SNR for Coil B allows small defects to be detected. The 2D maps of the UT and ASEM signal intensity at a magnetic field of 100 kA/m are shown in Figs. 11 and 12, where the white regions indicate a higher signal intensity. The most striking feature is that a through hole 0.1 mm in diameter is clearly detected in the ASEM map although it is not observed in the UT map.
2D maps of UT (top) and ASEM (bottom) signal intensities measured by using Coil B at 100 kA/m for a through hole (a) 0.5, (b) 0.3, and (c) 0.1 mm in diameter. The arrows indicate the positions of the through holes.
ASEM mapping around a through hole 0.1 mm in diameter. (a) 2D map and (b) 1D profile. The dotted line in (a) indicates the position of the 1D profile.
We discuss the pattern of the ASEM maps around holes. The ASEM signal intensity is reduced on the top and bottom sides of the hole and increased on the left and right sides (bottom panel of Fig. 11(a)). This pattern can be explained by the distribution of the magnetic flux density. Figure 13 shows the calculated distribution of the magnetic flux density around the hole at 100 kA/m. The magnetic flux density is also increased on the top and bottom sides of the hole and decreased on the left and right sides. At an applied magnetic field of 100 kA/m, the field dependence of the ASEM signal intensity has a negative slope (Fig. 4). Therefore, the contrast in the ASEM map is reversed for the distribution of magnetic flux density. This explanation is confirmed quantitatively. We focus on the low flux density area (black area in Fig. 13(a)) around a hole 0.5 mm in diameter. The magnetic flux density in the black area is roughly half that at a point away from the hole (1D profile along the X axis of Fig. 13(c)). The magnitude of the magnetic field then corresponds to about 50 kA/m. This indicates that the magnetic state in the black area is in the high-permeability region shown in Fig. 4. The ASEM signal intensity in the same area is about 1.8 V, which agrees with the value in the high-permeability region. However, the decrease in magnetic flux density on the left and right sides of a hole 0.1 mm in diameter almost recovered about 0.5 mm from the hole (Fig. 13(d)). The reduction of the contrast in the ASEM map for a hole 0.1 mm can be explained by the size of the ultrasound focal spot (1 mm in diameter).
Numerical calculation of magnetic flux density around through holes (a) 0.5 and (b) 0.1 mm in diameter at a magnetic field of 100 kA/m. (c) and (d) 1D profiles of magnetic flux density across the holes 0.5 and 0.1 mm in diameter, respectively. (Online version in color.)
Contrast in the ASEM map could be caused by changes in magnetism due to residual stress or variations in chemical composition. The contrast due to these factors will be lower than that caused by defects (see Fig. 5(a) in Ref. 4). The signal intensity may change significantly due to plastic deformation, such as indentations or scratches, but these are a type of defect. Variations in chemical composition usually occur on the scale of the grain size. Because the grain size in the thin sheet is typically 10–30 μm, which is much smaller than the size of the ultrasound focal spot, variations in chemical composition should not produce contrast in the ASEM map if the steel sheet is of sufficient quality.
In the ASEM wave technique, the spatial resolution is not affected by lift-off conditions. Thus, a small hole 0.1 mm in diameter can be detected even with a 10 mm lift-off. We also performed MFLT for a thin steel sheet with a through hole 0.1 mm in diameter (Fig. 14). Identifying a hole 0.1 mm in diameter is difficult for a lift-off of 2 mm in MFLT. Thus, the results confirm the sensor in the ASEM wave technique can handle a larger lift-off. Acoustical coupling to the objects is required for ultrasound irradiation, although the lift-off for ultrasound irradiation can be extended by the water immersion method3) or the laser ultrasound technique.
2D maps and 1D profile of the MFLT in a thin steel sheet with a hole 0.1 mm in diameter. The lift-off of the magnetic sensor (Hall sensor) is set to be (a) 0.5 and (b) 2 mm. In (c), the signal is measured through a high-pass filter for a lift-off of 2 mm. The measurements were performed for a horizontal leakage flux density of 0.06 T at a distance of 1 mm from the surface of the steel. The dotted line in the upper panels indicates the position of the 1D profiles.
We have investigated using the ASEM wave technique to detect flaws in thin steel sheets. By improving the SNR with a small coil antenna, we detected a through hole 0.1 mm in diameter with a 10 mm lift-off. Because the measurement speed of the ASEM wave technique is about 10 points per second, further improvements in the sensitivity and data acquisition speed will be required for its practical use in online inspection.