WO2018189498A1

WO2018189498A1 - Electromagnetic acoustic transducer based receiver

Info

Publication number: WO2018189498A1
Application number: PCT/GB2017/052869
Authority: WO
Inventors: Rachel Edwards; Oksana Trushkevych
Original assignee: The University Of Warwick
Priority date: 2017-04-11
Filing date: 2017-09-26
Publication date: 2018-10-18
Also published as: GB2561416A; GB2561551A; GB201715539D0; GB201705837D0

Abstract

An electromagnetic acoustic transducer (EMAT) based receiver is disclosed. The EMAT-based receiver comprises a permanent magnet (20) which extends along a first axis (21) between first and second ends (221; 222) and which is magnetised along a second, transverse axis (25) and first and second coils (281, 282) wound around the permanent magnet. The first and second coils are wound around the first axis and are spaced apart along the first axis. The first and second coils are wound close to the first and second ends respectively such that, when the receiver is presented to a sample (2), the first and second coils are located next to first and second surface regions (311, 312) of the sample in which magnetic flux (27) has first and second non-zero components along the first and second axes respectively.

Description

Electromagnetic acoustic transducer based receiver

Field of the Invention

The present invention relates to an electromagnetic acoustic transducer (EMAT) based receiver for sensing surface acoustic waves (SAWs) in sample and to a sensor which includes a transmitter from generating SAWs in a sample (which may be an EMAT- based transmitter) and an EMAT-based receiver.

Background

Inspection of surface-breaking defects is important in a number of applications such as, for example, monitoring integrity of rails or engine blades. To measure the geometry of a defect, phased array techniques can be used and they generally provide excellent imaging results. However, phased array techniques involve making contact with the sample under test and using couplant fill air gaps. If there is a coating on the sample, it must be removed prior to testing. Moreover, phased array techniques tend to be expensive and require complicated setup (particularly if generating Rayleigh waves), as well as complex analysis routines to fully optimise the results.

Often, however, an operator may only have access to the side containing the defects and so techniques have been developed for use in these circumstances, such as the so-called "satellite pulse" technique, whereby a shear wave is bounced off the far side of the sample and reflections back along the same path from any defects are detected. This can measure defects as small as 5% through-wall thickness. A near side detection and sizing transducer looks for reflections of a longitudinal wave propagating at a very shallow angle to the surface. Although this arrangement overcomes some of the issues of same-side inspection, it requires a complicated set-up.

The use of surface waves for inspection can make crack characterisation much simpler. Although electromagnetic acoustic transducers (EMATs) are less efficient transducers, they overcome many of these issues. For example, EMATs can be used when there are thin, electrically insulating coatings.

Various EMAT designs have been considered to optimise generation and detection of particular wave types, such as bulk, shear, and SAW, to enhance the capabilities of non- contact non-destructive testing. Linear coil designs are predominantly used for detection. More complicated designs have been developed, such as a butterfly-shaped detection coil and a double coil on a Halbach magnet for pipe monitoring using shear waves. Different shapes of coils, such as pancake-, meander-, rectangular- and racetrack-shaped coils, different magnetic field configurations and the use of pulsed magnetic fields have been considered for defect monitoring and thickness gauging. A combination of laser generation and EMAT detection has been used to detect in-plane components of a variety of wavemodes, such as shear, Lamb, longitudinal and mode- converted waves. EMAT designs for detecting in-plane or out-of-plane components of ultrasonic waves are becoming more widely known due to recent developments in EMAT design. These arrangements involve using two separate detectors which makes it is difficult to ensure that signals are correctly identified as being from the same defect. Processing methods for defect detection and thickness gauging are usually based on generating an ultrasonic wave and looking for reflections from imperfections in the sample such as cracks, voids and other defects. Recently there has been interest in the use of SAWs, in particular Rayleigh waves, for analysis of surface-breaking defects. These types of defects are challenging to detect and characterise using contact ultrasonic transducers. However, transmission of the waves underneath defects can be used to characterise the extent of the cracking into the material.

An example of this approach involves generating Rayleigh or Lamb waves on a sample and analysing reflected and transmitted waves. By using a broadband pulse, the depth of a surface-breaking defect that penetrates into a material can be measured.

Reference is made to R. S. Edwards, S. Dixon and X. Jian: "Depth gauging of defects using low frequency wideband Rayleigh waves", Ultrasonics, volume 44, pages 93-98 (2006). Constructive interference of incident waves and those resulting from reflection and mode-conversion at a defect can be used as a clear indicator of the location of a defect. The behaviour of waves in the vicinity of a defect depends strongly on the defect geometry. By considering this and relative behaviours of in-plane and out-of-plane wave velocity, some features of the geometry of the defect can be identified. Reference is made to M. H. Rosli, B. Dutton and R. S. Edwards: "In-plane and out-of-plane measurements of Rayleigh waves using EMATs for characterising surface cracks", NDT&E International, volume 49, pages 1 to 9 (2012), J. F. Hernandez-Valle, R. S. Edwards, B. Dutton, A. R. Clough and M. H. Rosli: "Laser ultrasonic characterisation of branched surface-breaking defects", NDT&E International, volume 68, pages 113 to 119 (2014) and R. S. Edwards, B. Dutton, A. R. Clough and M. H. Rosli "Enhancement of ultrasonic surface waves at wedge tips and angled defects", Applied Physics Letters, volume 99, page 094104 (2011).

Summary

According to a first aspect of the present invention there is provided an electromagnetic acoustic transducer based receiver comprising a permanent magnet which extends between first and second ends along a first axis and which is magnetised along a second axis which is transverse (for example, perpendicular) to the first axis and first and second coils wound around the permanent magnet. The first and second coils are wound around the first axis and are spaced apart along the first axis. The first and second coils are wound close (or "proximate") to the first and second ends respectively such that, when the receiver is presented to a sample (e.g. is placed next to the sample), the first and second coils are located next to first and second surface regions of the sample in which magnetic flux has first and second non-zero components along the first and second axes respectively.

The receiver is capable of extracting in-plane and out-of-plane components of surface acoustic waves in a sample disposed next to the receiver along the second axis.

The permanent magnet may comprise first and second opposite faces (or "poles") running between the first and second ends and through which magnetic flux passes. Thus, when a face (e.g. the first face) of the permanent is presented to the sample, magnetic flux passes into a surface region of the sample and the magnetic flux in first and second regions of the surface regions which lie proximate to the first and ends have first and second components along the first and second axes respectively.

The first and second coils may be contained within respective first and second end sections of the magnet lying at the first and second ends of the magnet respectively and extending no more than 4 mm from the first and second ends and preferably no more than 3 mm from the first and second ends. The coils are preferably separated by at least 1 mm, i.e. there is a gap of at least 1 mm between inner edges of the coil. The receiver may comprise at least one additional permanent magnet, preferably two additional permanent magnets, stacked on top of the permanent magnet along the second axis.

According to a second aspect of the present invention there is provided a sensor comprising a transmitter for generating ultrasonic surface acoustic waves in a sample and a receiver according to claim 1 or 2, spaced apart from the transmitter along the sample along the first axis, for measuring surface acoustic waves in the sample.

The transmitter may be an electromagnetic acoustic transducer. The transmitter electromagnetic acoustic transducer may be a racetrack-type electromagnetic acoustic transducer. Alternatively, other arrangements for generating waves can used, such as other electromagnetic acoustic transducer designs, a laser ultrasound source or a piezoelectric transducer. The receiver may be a first receiver and the sensor may further comprise a second receiver according to the first aspect of the present invention and the transmitter may be interposed between the first and second receivers.

This can help to reduce the need for a second scan if there are multiple or inclined cracks.

The sensor may comprise at least two rows arranged along a third axis which is transverse (for example, perpendicular) to the first and second axes. Each row may comprise a transmitter and a receiver. Each row may comprise a transmitter and first and second receivers wherein the transmitter is interposed between the first and second receivers.

This can help to cover a larger area while scanning and/or to reduce the need for a second scan if there are multiple or inclined cracks. According to a third aspect of the present invention there is provided a measurement system comprising a pulse generator configured to cause the transmitter to generate the surface acoustic waves in a sample, a sensor according to the second aspect of the present invention, wherein the first and second coils generate first and second time varying signals respectively and a signal processing system configured to generate first and second compensated time-varying signals by applying a time shift to at least one of the first and second time-varying signals respectively, to calculate an in-plane signal by summing the first and second compensated time-varying signals, and to calculate an out-of-place signal taking a difference between the first and second compensated time- varying signals. The signal processing system maybe further configured to display and/or to store the in-plane signal and/or out-of-plane signal.

According to a fourth aspect of the present invention there is provided a method of processing surface acoustic wave signals comprising receiving first and second time varying signals, generating first and second compensated time-varying signals by applying a time shift to at least one of the first and second time-varying signals respectively, calculating an in-plane signal by summing the first and second compensated time-varying signals and calculating an out-of-place signal taking a difference between the first and second compensated time-varying signals.

According to a fifth aspect of the present invention there is provided a computer program which, when executed by a computer, causes the computer to perform a method according to the fourth aspect.

According to a sixth aspect of the present invention there is provided a computer program product comprising a computer-readable medium, for example a non- transitory computer-readable medium, storing a computer program according to the fifth aspect.

Brief Description of the Drawings

Certain embodiments of the present invention will now be described, by way of example, with reference to the accompanying drawings, in which:

Figure 1 is a schematic block diagram of an inspection system which includes a sensor for generating and detecting surface acoustic waves in a sample and which includes an EMAT-based receiver, and a computer system for analysing detected signals;

Figure 2 is a perspective view of an EMAT-based receiver when presented to a sample; Figure 3 is side view of the EMAT-based receiver shown in Figure 2;

Figure 4 is a side view of a modified EMAT-based receiver;

Figure 5 is a schematic block diagram of a computer system;

Figure 6 is process flow diagram of a method of inspection;

Figure 7 shows a graph of measured signal amplitudes and a time-shifted measured signal amplitude plotted against time for a Rayleigh wave propagating through aluminium;

Figure 8 shows a graph of measured and calculated in-plane and out-of-plane signal amplitudes plotted against time for a Rayleigh wave propagating through aluminium; Figure 9 shows a graph of measured and calculated in-plane and out-of-plane signal amplitudes plotted against time for an So Lamb wave propagating through aluminium; and

Figure 10 shows a graph of measured and calculated in-plane and out-of-plane signal amplitudes plotted against time for an Ao Lamb wave propagating through aluminium; Figure 11 is a schematic block diagram of a sensor comprising an array of transducers for generating ultrasonic surface acoustic waves and an array of EMAT-based receivers; and

Figure 12 is process flow diagram of a simplified method of inspection. Detailed Description of Certain Embodiments

Referring to Figure 1, an ultrasonic system 1 for inspecting an electrically-conductive sample 2, such as rail or engine blade, using ultrasonic surface acoustic waves 3 is shown.

The system 1 can be used to inspect a surface 4 of the sample 2 to identify, locate and/ or characterise a crack 5, for example, in terms of penetration depth, d, and angle of the crack to the surface, Θ. The sample surface 4 may be covered by an electrically- insulating coating 5, such as a coat of paint or plastic. The system l includes an ultrasonic sensor 6 mounted to a positioning system 7, such as a motor-driven x-y stage, for moving the sensor 6 with respect to the sample 2. The ultrasonic sensor 6 comprises an ultrasonic transducer 8 (or "generator"), preferably in the form of an EMAT, for generating ultrasonic surface acoustic waves 2 in the sample 2 and a dual-coil EMAT-based receiver 9 for measuring ultrasonic surface acoustic waves 2 propagating in the sample 2. The generator 8 and receiver 9 are arranged in a line and are spaced apart along the line by a distance, s, in a range between 30 and 250 mm. In this case, the generator 8 and receiver 9 are spaced by 50 mm. The ultrasonic sensor 6 can be withdrawably presented to the surface 4 of the sample 2 to inspect the sample 2 and can be moved over the surface 4 of the sample using the positioning system 7.

The system 1 includes a pulse generator 10 for driving the ultrasonic transducer 8 at a frequency in a range 50 kHz to 2 MHz. The system 1 includes amplifiers 11 for amplifying time-varying signals received from the dual-coil EMAT-based receiver 9, an acquisition device 12, for example in the form of an analogue-to-digital converter (ADC) or digital storage oscilloscope (DSO), for digitising the received signals received from the amplifier 11, and a computer system 13 for processing the received signals received from the acquisition device 12. The acquisition device 12 may be incorporated into the computer system 13, e.g. in the form of an ADC or DSO card or module.

Referring to Figures 2 and 3, the dual-coil EMAT-based receiver 9 comprises a box- shaped (or "rectangular cuboid") permanent magnet 20. The permanent magnet 20 consists of neodymium-iron-boron (NdFeB) or other suitable magnetic material having a high remanent magnetization (e.g. compared to hard ferrite materials). The permanent magnet 20 extends along a first axis 21 (parallel to the x-axis) between first and second ends 22_1; 22₂. The permanent magnet 20 also extends along a second axis (parallel to the y-axis) between first and second sides 24! (the second side is not shown). Along a third axis 25 (parallel to the z-axis), the permanent magnet 20 lies between first and second faces 261, 262 (herein referred to as "lower and upper faces" respectively). The permanent magnet 20 is square in plan view and has a length, I of 8 mm, a width, w, of 8 mm and a thickness, h, of 4 mm. The permanent magnet 20 can have other geometries, aspect ratios and/or dimensions. The permanent magnet 20 is magnetized along the axis 25 parallel to the z-axis which is normal to the lower and upper faces 261, 262. Thus, when the lower face 261 of the permanent magnet 20 placed on or against a sample 2, magnetic flux 27 passes into the sample 2. The first and second fasces 261, 262 may also be referred to as "first and second poles".

First and second linear coils 281, 282 are wound around the permanent magnet 20, around the first axis 21. The coils 281, 282 may be wound onto a coil former (not shown) and the permanent magnet 20 maybe inserted into the former (not shown). The first and second coils 281, 282 are spaced apart along the permanent magnet 20, along the first axis 21, along and in line with the line that the generator 8 and receiver 9 are spaced.

The first and second coils 281, 282 are disposed close to the first and second ends 22_1; 22₂ respectively of the permanent magnet 20. Each coil 281, 282 comprises n turns of wire, and has a length, l_w, along the first axis 21 and a width, w_w, along the second axis 23, and are offset from the ends 22_1; 22₂ of the permanent magnet 20 by a distance, l_g. In this case, n = 8, l_w = 1.2 mm, w_w = 8 mm and l_g = 1 or 0.5 mm. The first and second coils 281, 282 are separated by a distance, l_s. The coils 281, 282 are placed the same distance from the ends 22!, 22₂ of the permanent magnet 20. In other words, the coil positions are symmetrical about a centre point between the two ends of the permanent magnet 20. Each coil 281, 282 has an inner coil edge 291 , 292 and an outer coil edge 29_1;0, 292 The inner edge 291 , 292 lies further away from the end of the magnet (i.e. the end pf the magnet closest to the coil) than the outer edge 29 292

The offset, l_g, preferably lies in the range o mm≤ l_g≤ 1 mm and more preferably o mm≤ lg≤ 0.5 mm. The length of the coil, l_w, depends on the frequency of the signal and can be chosen to obtain signal with sufficient signal-to-noise ratio. The width of the coil Ww can be chosen to obtain signal with sufficient signal-to-noise ratio and to be as small as possible to obtain sufficient spatial resolution. The coils 281, 282 are arranged

Referring also to Figure 4, one, two or more additional permanent magnets 20' having the same or similar geometry and dimensions as the permanent magnet 20 may be stacked on top of the permanent magnet 20 to form a stack 30. The magnetisations of the additional magnets 20' are orientated in the same direction as the permanent magnet 20. This arrangement can be used to increase magnetic field strength and reduce fringing fields. The coils 281, 282, however, are preferably not wound around the additional permanent magnets 20. In other words, the coils 281, 282 are only wound around the permanent magnet 20 at the base of the stack 30. This arrangement can help to reduce coil impedance and the associated so-called "dead time" of the amplifier 11 (Figure 1).

Referring again to Figures 2 and 3, when the sensor 6 (Figure 1) is placed on or against the surface 4 of the sample 2, magnetic flux 27 (herein also referred to as "B field") from the permanent magnet 20 enters into the sample 2 under the permanent magnet 20. The magnetic field 27 in the sample 2 has two magnetic field components, namely B_z

As explained earlier, the coils 281, 282 are wound close (or "proximate") to the ends 22_1; 22₂ of the permanent magnet 20. In particular, each coil 22_1; 22₂ is positioned over first and second surface regions 311, 312 of the sample 2 in which the magnetic flux 27 has sufficiently large, non-zero first and second components along the first and second axes, i.e. sufficiently large B_z and B_x. The relative values of the field components B_z and Βχ may be expressed as aB₀ and (ι-α)Β₀ where a lies between o and 1. Thus, a may take a value of between 0.2 and 0.8 (i.e. 0.2≤ a≤ 0.8) or preferably between 0.4 and 0.6 (i.e. 0.4≤ a≤ 0.6). In other words, the magnitudes of the components are similar or preferably about the same.

This can be achieved by positioning each coil 281, 282 such that the coil 281, 282 is contained within a respective section 321, 32₂ of the magnet 20 which extends from a respective end 22_1; 22₂ of the magnet. Herein, the sections 32_1; 322 are referred to as "end sections" of the magnet 20. The end sections 32_1; 322 may extend no more than 4 mm, and preferably no more than 3 mm, towards the middle of the magnet 20 from the ends 22_1; 22₂ of the magnet 20. Each coil 281, 282 is contained within a respective end section 32^ 322, but there is a sufficient gap between the inner edge 29^, 292 of each coil 281, 282. The coils are separated by at least 1 mm. In other words, there is a gap of at least 1 mm between inner edges 29^, 292 of the coil. Thus, (l_g+l_w)≤ 4mm, l_g≤ 1 mm and l_s≥ 1 mm. The receiver 9 employs EMAT detection using the Lorentz force mechanism which involves generation of a current resulting from charged particles moving in a magnetic field 27. The direction of the magnetic field 27 and the orientation of the coil 281, 282 determine which velocity component(s) is/are detected. Thus, for the dual-coil EMAT- based receiver 9, a field in the z-direction results in detection of in-plane velocities and a field in the x-direction results in detection of our-of-plane velocities.

First and second time-varying signal amplitudes A_t (t), A₂(t) are measured for the first and second coils 281, 282 respectively. Each signal contains a mix of the in-plane and out-of-plane components due to the presence of the two magnetic field components. As the x components of the magnetic field 27 are in opposite directions for the first and second coils 281, 282, this allows in-plane and out-of-plane velocity components to be extracted from the two signals.

Referring to Figure 5, signal analysis and control is implemented in software on the computer system 13. The computer system 13 includes at least one processor 33, memory 34 and an input/ output (1/ O) interface 35 operatively connected by a bus 36. The I/O interface 35 is operatively connected to storage 37 in the form of a solid state and/ or hard disc drive or drives, an interface 38 to removable storage 39 (such as a flash drive), one or more network interfaces 40. The computer system 13 also includes one or more displays 41 and one or more user input devices 42 (such as keyboards, mouse and/or a touch screen). Data processing software 43 and control software 44 are stored on the hard drive 37.

Referring to Figures 1, 2, 3 and 6, a method of signal processing will now be described. The sensor 6 is placed on or against the surface 4 of the sample 2. The computer system 13, automatically or under user control, causes the positing system 7 to position the sensor 6 at a position ; (step Si). The computer system 13 receives signals (that is signals from the coils 281, 282 via the amplifier 11 and acquisition device 12) for first and second coils 281, 282 (step S2).

Before in-plane and out-of-plane components of the signals can be extracted, a time delay, At, is introduced into one of the signals because the wave 3 propagating in the sample 2 will arrive at the first and second coils 281, 282 at different times due being separated by the distance, l_s (step S3). The time delay, At, depends on the spatial coil separation, l_s, as well as on wave propagation velocity (which depends on the sample material and geometry). A calibration step, to determine the time delay, need only be performed once, namely after the coils 281, 282 are manufactured. A measurement using ultrasound behaviour can provide a more accurate adjustment than measuring the coil separation using a ruler.

A time delay, At, that is necessary to achieve temporal overlap of the first and second signals is obtained using a configuration where the magnetic field is the same for both coils. In the case where the detection coils 281, 282 are wound on the coil former (not shown) such that the coils 281, 282 can be removed from the magnet, this is achieved by inserting a magnet (not shown) into the coil former (not shown), ensuring that the magnet is much larger (i.e. longer) than the coil separation such that the magnetic field at each coil can be taken to be the same. The magnetic field is directed into the sample, giving the in-plane velocity component. In the case where the detection coils 281, 282 are wound directly onto the permanent magnet 20, additional magnets (not shown) maybe placed either side of the permanent magnet 20 to reduce the horizontal magnetic field component in the sample 2. If this calibration is done on the same material and using the same ultrasonic wavemode which will be used for future tests (e.g. Rayleigh wave on a thick sample of aluminium), no further actions are necessary, and the measured time delay can be used directly in the calculations. If the receiver 9 is to be used on a variety of different materials and/or using Rayleigh and/or Lamb waves, then speed of the particular test wave is used to calculate the spatial coil separation. This can then be converted back to time delay for other materials and/or other wavemodes. The wave velocity for the wavemode used for testing is measured with high precision by doing time-of-flight measurements at varying distances from the generating coil. Figure 7 shows a Rayleigh wave in an aluminium sample measured by the two detection coils 281, 282, and the effect of a time delay of At = 1.85 μβ. This time delay corresponds to a coil separation, l_s, (Figure 2) of 5.4 mm.

Once the time delay has been factored in, the in-plane and our-of-plane components of the signals are calculated by taking the sum and differences respectively of the two signals, namely:

where Ai and A₂ are the amplitudes of the first and second time-varying signals suitably corrected using the time shift (step S6).

The computer system 13 applies a geometric shift (step S7) and the in-plane and out-of- plane components using equations (1) and (2) above (step S8).

Figure 8 shows the result extracting the in-plane and out-of-plane components using equations (1) and (2) above. In Figure 8, the result is labelled 'calculated from differential' and is compared measurements of in-plane and our-of-plane components measured using separate EMATs (not shown) arranged to detected in-plane or out-of- plane velocity components.

Figure 8 shows an S₀ Lamb wave mode for a frequency-thickness of 0.3 MHz.mm, measured on an aluminium plate. This mode should be substantially non-dispersive for the frequency bandwidth generated. The majority of the energy of the S₀ wavemode at this frequency-thickness is in in-plane motion and so the out-of-plane signals have been scaled by a factor of 2 to show the shape of the wave.

Figure 9 shows the A₀ Lamb wave mode, with the time delay calculated for a frequency- thickness of 0.15 MHz.mm. The dashed lines show the wave arrival time window for this frequency-thickness, identified using a sonogram. This mode is expected to be dispersive, hence different delays are required for different frequency-thicknesses. However, calculation of the time difference at a central frequency will enable identification of the chosen mode behaviour. For both Lamb wavemodes, the amplitude distribution between the in-plane and out-of-plane components is as expected.

Referring still to Figures 1, 2, 3 and 6, the computer system 13 builds an amplitude map for the in-plane and out-of-plane components (step S9).

The computer system 13 may, optionally, scan in the reverse direction (step S10). Thus, if a first scan is performed by moving the ultrasonic sensor 6 from a first position A to a second position B, a second scan, in the reverse direction, is performed my moving the sensor 6 from the second position B to the first position A. For each scan, the receiver 9 is at the front.

The computer system 13 calculates transmission, enhancements, enhancement ratio (step S11), obtains crack position (step S12), obtains defect angle (step S13) and defect depth (step S14).

For characterising cracks, transmission coefficients can give a measure of the depth of the crack by looking at how much signal is blocked by a crack or which frequencies are blocked, while signal enhancement measured for the IP and OP velocity components when the EMAT is above a defect can give the angle of propagation into the material. This enhancement is due to interference between incident and reflected waves. For both of these measurements, an initial reference signal is measured on a section of sample which does not contain defects to give a comparison amplitude which can be compared to wave amplitudes during the scan when either enhanced or reduced due to a defect. Using this process, the enhancement coefficients can be obtained from a single scan, without positioning errors, by finding and storing signals with maximum amplitudes from both coils and correcting for the separation. Enhancement factors can then be calculated as described in R. S. Edwards, S. Dixon and X. Jian: "Depth gauging of defects using low frequency wideband Rayleigh waves", ibid.. If liftoff of the receiver 9 above the sample 2 can be kept constant during the scan, then only one reference calculation is required at the start of the scan to give the signal amplitude over a section with no defects. Where liftoff is not constant, in-plane and out-of-plane reference values are calculated for each scan point.

Referring also Figure 11, a sensor 6A is shown which comprises a row (along the y-axis) of M generators 8 and a corresponding row (along the y-axis) of N receivers 9, where M ≤ N and the generator 8 width * M≥ receiver 9 width * N, and where N≥ 2 and preferably 5≥ N≥ 20.

Figure 12 shows a simplified method of signal processing.

Referring to Figures 1, 2, 3 and 12, the user positions the sensor 9 away from the crack 5, instructs the computer system 13 to acquire reference, and the computer system 13 acquires waveforms which serve as a reference (step S15). The user then positions the sensor 6 over the crack in a way that the crack 5 is between the transmitter 8 and receiver 9, and instructs the computer system 13 to acquire transmission. The computer system 13 acquires waveforms transmitted through the crack (step S16).

The user positions the sensor 6, approaching the crack from one chosen side so that a maximum amplitude of enhanced signal is obtained from the first coil 281. The user instructs the computer system to acquire maximum amplitude on coil 281. The computer system 13 acquires a first waveform from the first coil 281 (step S17). The user then moves the sensor 6 to obtain maximum enhanced signal on the second coil 282, instructs the computer system to acquire maximum amplitude on coil 282. The computer system acquires a waveform from the second coil 282 (step S18).

Steps S15, S16, S17 and S18 maybe performed in any order.

The user then may, optionally, use the system to acquire transmission and enhanced signals approaching the crack from an opposite side, which is equivalent to scanning in reverse direction (step S19).

The computer system 13 calculates transmission, enhancements, enhancement ratio (step S20), obtains crack position (step S21), obtains defect angle (step S22) and defect depth (step S23).

Scanning in forward and reverse directions, acquiring signals and/or processing signals maybe carried out by the computer system automatically, i.e. without user

intervention.

The sensor may include two rows of receivers 9 with a generator 8 interposed between rear and front receivers 9 and each set of one generator 8 and two receivers 9 arranged in a line (along the x axis). It will be appreciated that various modifications may be made to the embodiments hereinbefore described. Such modifications may involve equivalent and other features which are already known in the design, manufacture and use of EMATs and NDA systems and component parts thereof and which maybe used instead of or in addition to features already described herein. Features of one embodiment may be replaced or supplemented by features of another embodiment. More than one permanent magnet maybe used. For example, additional magnets may be placed, along the first axis, either end of the permanent magnet.

Although claims have been formulated in this application to particular combinations of features, it should be understood that the scope of the disclosure of the present invention also includes any novel features or any novel combination of features disclosed herein either explicitly or implicitly or any generalization thereof, whether or not it relates to the same invention as presently claimed in any claim and whether or not it mitigates any or all of the same technical problems as does the present invention. The applicants hereby give notice that new claims may be formulated to such features and/ or combinations of such features during the prosecution of the present application or of any further application derived therefrom.

Claims

1. An electromagnetic acoustic transducer based receiver comprising:

a permanent magnet which extends between first and second ends along a first axis and which is magnetised along a second, transverse axis; and

first and second coils wound around the permanent magnet wherein the first and second coils are wound around the first axis and are spaced apart along the first axis, wherein the first and second coils are wound close to the first and second ends respectively such that, when the receiver is presented to a sample, the first and second coils are located adjacent to first and second surface regions of the sample in which magnetic flux has first and second non-zero components along the first and second axes respectively.

2. A receiver according to claim 1, comprising at least one additional permanent magnet stacked on top of the permanent magnet along the second axis.

3. A sensor comprising:

a transmitter for generating ultrasonic surface acoustic waves in a sample; and a receiver according to claim 1 or 2, spaced apart from the transmitter along the sample along the first axis, for measuring surface acoustic waves in the sample.

4. A sensor according to claim 3 wherein the receiver is a first receiver, wherein the sensor further comprises a second receiver according to claim 1 or 2 and wherein the transmitter is interposed between the first and second receivers.

5. A sensor according to claim 3, comprising at least two rows arranged along a third axis which is transverse to the first and second axes, each row comprising a transmitter and a receiver.

6. A sensor according to claim 4, comprising at least two rows arranged along a third axis which is transverse to the first and second axes, each row comprising a transmitter and first and second receivers wherein the transmitter is interposed between the first and second receivers.

7. A measurement system comprising: a pulse generator configured to cause the transmitter to generate the surface acoustic waves in a sample;

a receiver according to claim 1 or 2, wherein the first and second coils generate first and second time varying signals respectively; and

a signal processing system configured:

to generate first and second compensated time-varying signals by applying a time shift to at least one of the first and second time-varying signals respectively; and to calculate an in-plane signal by summing the first and second compensated time-varying signals; and

to calculate an out-of-place signal taking a difference between the first and second compensated time-varying signals.

8. A measurement system according to claim 7, wherein the signal processing system is further configured to display the in-plane signal and/or out-of-plane signal.

9. A measurement system according to claim 7 or 8, wherein the signal processing system is further configured to store the in-plane signal and/or out-of-plane signal.

10. A method of processing surface acoustic wave signals comprising:

receiving first and second time varying signals;

generating first and second compensated time-varying signals by applying a time shift to at least one of the first and second time-varying signals respectively;

calculating an in-plane signal by summing the first and second compensated time-varying signals; and

calculating an out-of-place signal taking a difference between the first and second compensated time-varying signals

11. A computer program which, when executed by a computer, causes the computer to perform a method according to claim 10.

12. A computer program product comprising a computer-readable medium storing a computer program according to claim 11.