Introduction

Longitudinal waves and vertically polarized Shear waves (S_v - Waves) have been used for some time now in ultrasonic testing. These two kinds of spherical waves can be generated by using a coupling medium between the piezoelectric transducer and the work piece. The third kind of spherical wave - horizontally polarized Shear waves (S_H - Waves) - are not often applied in ultrasonic testing because until recently only conventional transducers, which require very viscous couplants, were used, making transducer manipulation virtually impossible.

Now electromagnetic ultrasonic transducers (EMAT) have been developed. These can generate and receive S_H - Waves [100, 107]. They don't need a couplant for transmitting waves in metals, just close contact with the surface. The following chapter briefly describes the different methods of generating electromagnetic S_H - Waves, EMAT transducers and some of their characteristics.

5.2 Principle of S_H - Wave - Generation with EMAT Transducers

EMAT transducers generate ultrasound waves in electric conductive materials by Lorentz force, by electro magnetostrictive effect, or by a combination thereof. Both phenomena affect the atomic grid of the work piece, so that the acoustic waves are generated directly within the work piece instead of inside the transducer as with piezoelectric transducers.

The Lorentz force is caused by interaction between the electric current (current density J) which is inducted through the eddy current coil and the magnetic flux (B₀). The direction and intensity of the force F_L is determined by the vector equation:

Fig 5.1: Principle of electromagnetic acoustic generation with Lorentz-force.	Fig 5.2: Principle of electromagnetic acoustic generation with Magnetostriction
Fig 5.3: Different kind of Radio Frequency (RF) coils
Fig 5.4: EMAT - Probes for SH - Waves with Electromagnet

F_L = J x B₀ (5.1)

The Lorentz force, electric current and magnetic field lie at a 90° angle from each other, Abb. 5.1.

Physical distortion occurs in almost all ferromagnetic materials when a magnetic field is applied. This phenomenon is called 'magnetostriction' (in simple words: it is the magnetic equivalent of the piezoelectric effect).

The distortion appears generally parallel to the applied magnetic field. If an eddy current coil is mounted on the surface of a ferromagnetic material (steel), in addition to the eddy current field, dynamic magnetic fields are inducted generating the dynamic forces F_ms. These forces lie parallel to the applied magnetic flux B_o, Fig. 5.2.

To achieve a sufficient sound beam it is generally necessary to use big apertures by means of big eddy current coils. Two concepts are available. A big aperture can be achieved by using a large scale meander coil or by an arrangement of many small coil segments, Fig 5.3. Such a transducer of multiple coil segments works like a phase controlled transducer array. The eddy current coils or the coil segments are operated by a frequency burst and alternating current. The burst length relies on the coil's size.

If the burst frequency and the sound speed of the generated wave mode is known, the wave length can be calculated. The distance _s between two elements of the eddy current coil of the same current direction is called track wavelength.

What are the advantages and disadvantages of both concepts: non-segmented and segmented coils (phased controlled arrays)? Big non-segment coils generating narrow band signals mean the axial resolution is bad. Furthermore, EMATs of non-segmented coils distribute the same energy forward and backward.

EMAT - Probes with segmented coil, which work as a phase controlled array, provide ultrasound signals of wider bandwidth than non-segmented probes and deliver a forward to backward ratio of 30 - 40 dB. Such electromagnetic, phase controlled arrays are better suited for ultrasonic testing than EMAT Probes with non-segmented coils.

5.2.1 EMAT - Probes for ferrite Material
Fig. 5.4 shows a top view of the contact surface of a EMAT Probe which is only used for ferromagnetic materials [105].

The U-shaped electromagnet provides a quasi static magnetic field which lies parallel to the material's surface. The magnet is operated by alternating current (e.g. 50 Hz) so it can be easily moved on the work piece. The eddy current coil provides multiple turns and its axis lies parallel to the magnetic field. The eddy current and the magnetic field propagate parallel to each other so that the magnetostrictive effect on ferromagnetic materials is dominant.

Such electromagnetic probes are built as non-segmented as well as segmented eddy current probes (principle of the phased control array), Fig. 5.4. The coils are driven by amplifiers which generating the burst signal. In case of segmented coils each coil is driven by a separate amplifier of the delayed burst signal.

EMAT probes are generally built as separate transmitters and receivers (eddy current coil), either both coils are interlocked or the coils are side by side in a V-shape.

With such a S_H-Wave Probe it is possible to achieve a dynamic range of 45 dB on ferrite steel. Typical probe frequencies are in the range of 0,6 - 1,5 MHz and typical track wave lengths are in the range of 3 - 5 mm. Phase controlled arrays (probes with segmented eddy current coils) deliver a forward-backward radio of typically 30 dB for angles of incidence in the range of 30° - 60°.

The sound beam divergence of an EMAT probe depends on the angle of incidence. Generally it is true that a wider angle of incidence equals a wider beam. A meander probe (transmitter probe) with 12 elements and a track wavelength of 4 mm for 45° angle of incidence (f = 1,15 MHz) typically provides +-7 mm beam width (at 6 dB drop). The increase of the angle of incidence to 60° and the frequency change to 0,95 MHz increases the beam divergence to +10°. An increase in the number of array elements reduces the beam divergence. A phased controlled array probe, for instance with 6 segments, each containing 3 array elements with 4mm track wave length provides a beam divergence of +-3° for an angle of incidence of 45°. The increase of the angle of incidence to 60° increases the beam divergence to +-5,5°.

5.2.2 EMAT - Probes for non- ferrite materials
EMAT probes generate S_H-waves in non- ferrite or low ferrite materials by the Lorentz-force phenomenon.

Fig 5.5: EMAT probe for SH-waves with Permanent Magnet

Fig. 5.5 (left part) shows a non-segmented or segmented EMAT Probe for S_H-waves with permanent magnet [105, 185].

In the case of segmented (phased controlled array) probes, an arrangement of permanent magnets (segments) is used to generate the magnetic bias field, as shown at the right of Fig. 5.5; the eddy current is generated with the meander coil.

The propagation direction of the S_H - wave, which is generated by the meander coil and a segment of single permanent magnets, is led directly into the material perpendicular to the plane of the permanent magnets.

If the meander coil is driven time delay controlled, the probe operates as a phased control array, and the angle of incidence of the S_H - wave can be controlled between 30° and 90°.

Fig 5.6: EMAT- SE- Probe, layout of permanent magnet Fig 5.7: Control of the angle of incidence with variation of the frequency (non-segmented EMAT- Probe)

A S_H - wave - EMAT - Probe usually consists of a separate transmitter and receiver probe, as described above. Fig 5.6 shows a typical layout of permanent magnets in a TR probe [107]. If this is applied on a fully austenitic material, generally a dynamic range of 35 dB can be achieved. The dynamic range can reach 45 dB on ferrite austenitic steels and austenitic cladding.

The track wave length _s of a S_H - wave - probe with permanent magnets is determined by the distance of two permanent magnets of the same polarity lying side by side. Typical values of the track wave length _s are 4 mm to 6 mm, and a frequency range of 0,8 MHz to 0,5 MHz. The probe bandwidth of 50 - 100% depends on the width of each element, and the forward-backward ratio is typically larger than 35 dB for separate transmitter-receiver probes. The sound beam divergence of a phased controlled array with six segments of 3 elements each and track wavelength of 5 mm is +-5,5°, for an angle of incidence of 60°.

These permanent probes were mainly developed for testing of non-ferrite materials. There they can be applied with no problems since they are easy to handle. On ferrite materials this probe can very difficult to move, because of its magnetic force. But with the help of special probe manipulators it is possible to use permanent magnetic probes on ferrite materials also.

5.2.3 Angle of Incidence and Frequency of EMAT - Probes
The wavelength () of the spherical wave which is generated by an EMAT - probe relies on the operation frequency (f) of the eddy current coils.

So it is easy to calculate = c/f, where c is the sound velocity of the generated wave mode.

The angle of incidence () of an EMAT- probe can be determined in the case of a non-segmented probe by the operation frequency and the track coil wavelength (_s) of the transducer [130], Fig. 5.7, accordingly the ratio:

sin = /_s (5.2)

Fig. 5.8 shows the angle of incidence of a probe as a function of the transducer's frequency and the track wavelength for generation of longitudinal - and shear waves.

Fig. 5.9 demonstrates that the main lobe of a non-segmental probe can be controlled by frequency variation [102]. Here the sound distribution of a transmitter at a halve cylinder of ferrite steel of 150 mm radius was measured.

Fig. 5.8 Incidence angle of a probe as a function of the transducer's frequency and the track wavelength [mm]. Fig 5.9: Sound distribution of a non-segmented EMAT probe (track wavelength: 4 mm, 12 array elements)

An EMAT probe with a meander coil was applied (Track wave length: 4 mm, 12 elements). If the transmitter is operated at app. 800 kHz, the maximum main lobe appears with a 90° angle of incidence.

The angle of incidence can change from 60° and 40° when the frequency varies between 950 kHz and 1,3 MHz. The level of side lobes is very low; that is based on an amplitude distribution of each coil element (variation of turns).

Fig 5.10: Control of incidence angle by variation of the delay time (segmented EMAT probe)

The sound beam divergence will be lower if the number of array elements increases. The angle of incidence of a phased array, Fig. 5.10, is mainly determined by the delay time (t) each element is driven with, the distance of the segments and the sound velocity (c):

sin = c t/d (5.3)

The frequency of the phased array is determined by the track wavelength of each element. If the number of elements within a segment is small (1-3), it is possible to control a large range of glancing incidence angle (90° - 50°) by varying the delay time.

That is possible since the main lobe of a segment with, for instance, 3 elements is very wide, Fig. 5.11 [103]. Within this wide main lobe of the single element it is possible to control the lobe of the complete array only by varying the delay time.

If a large number of array elements (> 3) are used for a segment, the frequency of the probe must be changed, if the delay time is changed, to achieve another angle of incidence.

Since the main lobe of the segments gets smaller when the number of array elements increase, for an unchanged frequency a loss of sensitivity outside the main lobe maximum is predictable.

Fig. 5.11 Sound distribution of one segment of a segmented EMAT probe. Fig 5.12: Sensitivity of an EMAT probe along the acoustic axis

5.3 EMAT- TR- Probes

5.3.1 Axial sensitivity ("Focus length")
Similar to conventional piezoelectric TR - probes, EMAT - TR - probes provide a specific distribution of axial sensitivity which relies on the V - position of the transmitter and receiver. Fig. 5.12, shows the sensitivity along the acoustic axis of an EMAT TR Probe of a 5 mm track width, 17 mm width of the transmitter and receiver and a V - position of 18°. The "focus point" lies at a distance of 80 mm [107].

Thickness of the work piece Range of incidence angle Dead zone of the probe Length and width of the probe => Maximum sensitivity at 95 mm "Focus length": 95 mm Fig 5.13: Parameters which determine the V- Position of an EMAT- TR- Probe

The V-position of transmitter and receiver of an EMAT probe can be optimized for a certain wall thickness and incidence angle as follows: Fig. 5.13: If the far-surface of a work piece of 40 mm is to inspect with an angle of incidence between 55° to 70°, it needs a focus distance of 95 mm because this distance is equal to the mean incidence angle of 55° to 70°.

5.3.2 Sound distribution perpendicular to the beam axis
The main lobe of an EMUS - TR - Probe with permanent magnets (four receiver and four transmitter elements), 17 mm receiver and transmitter width and a 18° V - position between transmitter and receiver is shown in Fig. 5.14. The 6 dB - width of the main lobe divergence is 15° [107].

5.3.3 Sound distribution in front of the probe


Fig. 5.14 Sound distribution of an EMAT - TR - Probe perpendicular to the beam axis (transmitter) Fig. 5.15 Distributions of an EMAT - TR - Probe in front of the probe.

Fig. 5.9 showed the sound distribution of EMAT - Transmitter Probes with non-segmented RF - coil, measured in front of the probe. The incidence angle was controlled by varying the frequency.

For EMAT - Probes with segmental RF coils (arrays) the angle of incidence is controlled by varying the delay time. Fig. 5.15 shows the distribution of a permanent magnetic probe with four transmitter and receiver segments (_s = 4,4mm, 4 elements per segment).

The distribution was measured at an Aluminum - half cylinder with a 10 mm radius. The EMAT - Probe was designed for inspection of piping of 324 mm outer diameter. The distributions shown in Fig. 5.15 were performed by using the following delay times:

Angle of incidence	t1 [ns]	t2 [ns]	t3 [ns]
60°	3054	5959	8763
70°	3288	6456	9545
80°	3414	6777	10111
90°	3456	6892	10318

The delay times t₁, t₂ and t₃ refer to the first segment of the transmitter probe.

Dr. rer. nat. Gerhard Hübschen
The author is principal investigator at
Fraunhofer Institut Zerstörungsfreie Prüfverfahren IZFP,
University Bldg 37, D-66123 Saarbrücken
Phone: +49 681 302 3840
Fax: +49 681 39580
Email: eschneider@izfp.fhg.de
Homepage: http://mm.fhg.de/depts/izfp-e.html
WebSite: On NDTnet

This article is the Chapter 5 of the Book: "Ultraschallprüfung von austenitischen Plattierungen, Mischnähten und austenitischen Schweißnähten" Eberhard Neumann et al Eberhard.Neumann@bam.de See also online the Chapter 4: Ultrasonic Testing of Austenitic and Dissimilar Metal Welds Expertverlag 1995 Kontakt & Studium; Bd. 377: Werkstoffe ISBN-3-8169-1078-5 Mit 340 Seiten, 204 Bildern und 210 Literaturstellen (c) Expert Verlag; D- 71272 Renningen Malmsheim. Homepage: www.expertverlag.de