акустика / gan_ws_acoustical_imaging_techniques_and_applications_for_en

Specimen surface

–z

θR

Focal plane

Figure 9.11 Ray model of an acoustic lens with negative defocus: aa is an arbitrary ray, which is reflected at such an angle that it misses the transducer, or hits the transducer obliquely and therefore contributes little to the signal because of phase cancellation across the wavefront; bb is the axial ray, which goes straight down and returns along the same path; cc is the symmetrical Rayleigh propagated wave, which returns to the transducer normally and so also contributes to the signal. The wavy arrow indicates the Rayleigh wave (Briggs [22])

182 Acoustical Imaging: Techniques and Applications for Engineers

the fluid at the Rayleigh angle. The particular ray of importance is the one that propagates back to the lens along a path symmetrical to the initial ray responsible for exciting the Rayleigh wave. Both the axial ray and the Rayleigh ray contribute to the signal at the transducer. These two rays are incident at different places on the transducer. The piezoelectric voltage they excite is summed with respect to both the amplitude and the phase. The complex-valued sum is then detected and, hence, the interference effects between them are observed [9, 10]

As z changes, the phase of these two rays changes at different rates, so that they will alternate between constructive and destructive interferences. The phase φ0 of the geometrically reflected

where k is the wavenumber of the fluid.

The phase φR of the Rayleigh wave advances by virtue of the shortening of the path in the fluid, but against this is the path of the Rayleigh wave in the specimen surface. Therefore the

with kR = k sin θR from Snell’s law.

The phase change of π in (9.2) is due to the phase change of π in the reflection coefficient at the angle.

If the output of the transducer is usually detected by a phase-insensitive circuit, then the difference between the phases of the two rays is important; that is

As the specimen is moved towards the lens, the two rays will alternate between being in phase and being out of phase. The period z of the resulting oscillations in V (z) is the movement in the z-direction needed for a change of 2π in the relative phase,

where Vw is the wave velocity in water and VR is the wave (or Rayleigh wave) velocity on the surface of the specimen.

The expression for the period of the oscillations in V (z) is of fundamental importance in the study of materials and in quantitative acoustical microscopy because one can obtain the Rayleigh velocity, VR, from this expression and VR can be used to derive the various elastic moduli such as Young’s modulus, bulk modulus and Poisson’s ratio.

9.4.3 V (z) Curve Technique in the Characterization of Kissing Bond

Adhesive bonding in an aircraft structure is an active area of R&D. Bonding can be used for structural joining and attachments in commercial aircraft. There is a long history of metal bonding in the primary load-bearing application for some small airplanes or jets. Extensive bonding is used in prop-driven airplanes, such as for composite sandwich skin panels and major joints to close the wing torque box and to attach main spars and fuselage skin splices. Business jets also use a bonded sandwich method in the fuselage, while major fuselage splices include a bolted redundancy. For rotorcraft and propellers, there is a combination of bolted and bonded structures in the airframe and in dynamic parts. Major splices are bolted and many are bonded attachments. For transport aircraft, there are bonded attachments (stringens, sandwich panels) for composites, but major joints remain bolted. Bond characterization and durability tests are critical for aircraft safety, but as yet bond properties are not measureable prior to fabrication and, in any case, there is no nondestructive test that can assure bond integrity. Generally, bonded repairs are not allowed in structurally critical applications. There is also an absence of a means of verifying bond strength in the finished product. There are two types of adhesive bounding problems [11]:

1.Kissing bond: A dry contact type making the contact of two compressed but otherwise unbounded surfaces; a wet contact type, meaning that the bond consists of two disbonded surfaces separated by a thin layer of liquid, such as contamination. It is difficult to detect kissing bonds using conventional nondestructive testing techniques such as pulse–echo ultrasound because the resulting change in stiffness is often very small.

2.Weak bond: The residual strength seen in weak adhesive bonds makes them significantly more challenging to detect than a kissing bond.

Today even the easier kissing bonds are still in the research stage (see Figures 9.12 and 9.13). Currently NDT techniques for delamination in composite parts with monolithic and sandwich construction is commonly through ultrasonic technique. The ultrasonic technique is able to detect voids between layers. However, if the bond strength has deteriorated between two layers but a void has not developed, the ultrasonic technique is unable to identify such a kissing bond.

To determine the bond strength, a pull test is performed. However, a pull test is destructive and loads the part until bond line failure occurs, and the part has to be replaced or repaired after the test. Also, current NDE methods only present an indication of contact between two bonded parts and not the load-carrying capability. So far, the ultrasonic work performed on kissing bonds is qualitative in nature, but no quantitative nature of the kissing bond is given.

9.5Dry Contact or Noncontact Transducers

The main weakness of pulse–echo ultrasound is that a couplant is needed to be applied between the specimen and the probe. Extra labour is involved, and if this couplant layer is not consistent, erroneous results can occur. Combination of the test specimen may also be a problem. The sonic pitch/catch and mechanical impedance procedures do not require a couplant. Another weakness of the ultrasound pulse echo technique is that it lacks sensitivity to near surface defects, particularly if a signal crystal probe is used. The signal from the surface of the specimen will then interfere with the signal from the near surface defect which produces a dead zone near the specimen surface. This problem is particularly acute in composites which are invariably thin. The problem can be overcome by using a through-transmission probe echo ultrasound, but this procedure then has the weakness of necessity from both specimen sides, a couplant requirement and no indication of defect depth. The sonic NDT procedures hence also have advantages over through-transmission ultrasound.

9.5.1Defect Depth, Sizing and Characterization

Detect the sensitivities of these procedures in terms of the smallest defect detectable at various depths. The defect characteristics are sensitivities in terms of the smallest defect detectable and the defect depth, size and type (i.e. delamination, disbond, porosity, etc.)

9.5.2Pitch/Catch Swept Method

A dual element point probe is used with no couplant. One element transmits sound through the specimen to the other probe. The frequency is swept and changes in the resulting frequency can detect the presence of defects.

9.5.3Pitch/Catch Impulse Method

A dual element point probe is also used with no couplant and a burst of sound energy of several cycles is transmitted from one element to the other. A time gate is utilized with different travel times to monitor the presence of defects.

9.5.4MIA Test Method

This mechanical impedance method uses a dual element with no couplant. The driver element generates a continuous sound wave into the specimen, which is then received by the receiving element. Differences in loading are used to detect the presence of defects. Its effectiveness is a combination of sensitivity in terms of the smallest defect detectable at specific depth and an ability to determine the depth, size and type. It is likely that all the sonic techniques will have better sensitivity to very near defects than pulse–echo ultrasound.

For dry contact or noncontact acoustical image, dry transducers have also been used. However, their frequencies are comparatively low, typically in the range of 600 kHz to 2 or 3 MHz. Ideally the ultrasound energy should be generated via an intermediate air gap which

may be varied according to the surface geometry and scanning applications. The air-coupled (noncontact) ultrasound method offers considerable potential for the rapid NDE of aircraft and aircraft components. For both metallic and nonmetallic structures, two techniques are of interest for production and in-service inspection. Laser-generated ultrasound, combined with laser detection, offers the significant advantage of truly remote testing with the potential for rapid scanning of complex geometrical components. Laser systems have been demonstrated for through transverse scanning [12], plate wave generation [13] and pseudo-array imaging [14]. However, the prototype systems are expensive, relatively cumbersome, require safety precautions and can cause damage to the surface of the inspection system. An alternative method involves conventional ultrasound scanning but with the liquid couplant replaced by an intermediate air gap. Such systems are relatively inexpensive and have the advantages that existing scanning methods (including manual operation) and an NDE standard may be used for inspection and assessment. A major difficulty is the 140 dB reduction in the available signal [15] when compared with a standard water-coupled system. This arises from air attenuation and mechanical mismatching between the transducer/air/test specimen interfaces. For reliable operation within an industrial environment that may be subjected to rough or uneven surfaces, humidity/temperature variations and draughts, a substantial amount of this degradation has to be circumvented. Air-coupled ultrasound scanning offers an attractive way forward for more rapid and versatile inspection. Indeed, the technology would complement a future laser scanning system for an overall noncontact NDE package.

The advantage of a noncontact NED system is its speed of scanning, especially for large structures like aerospace, nuclear, oil, gas, maritime and automobile industries.

9.6Phased Array Transducers

9.6.1Introduction

Phased array ultrasonic technology moved from the medical field to the industrial sector at the beginning of the 1980s [16, 17]. By the mid-1980s, piezo-composite materials became available for the manufacture of complex-shaped phased array probes [3, 4]. By the beginning of the 1990s, phased array technology was incorporated as a new nondestructive evaluation method in ultrasonic handbooks [5, 6] and training manuals for engineers [7, 8]. The majority of the applications from 1985 to 1992 were related to nuclear pressure vessels (nozzles), large forging shafts and low-pressure turbine components. The rapid advance in micromachining, microelectronics, piezo-composite technologies and computation power (including simulation packages for probe design and beam-component interaction), all contributed to the revolutionary development of phased array technology by the end of the 1990s. Functional software was also developed as computing capability increased.

Currently, phased array technology is used for nondestructive testing (NDT) by the following industries: petrochemical, power generation, aerospace, defence, and manufacturing. It is triggered by the following inspection requirements:

1.Decreased inspection time.

2.Detection of randomly orientated cracks at different depths using the same probe in a fixed position.

3.Improved signal-to-noise ratio (SNR) and sizing capability for dissimilar metal welds and centrifugal-cast stainless-steel welds.

4.Detection and sizing of small stress-corrosion cracks (SCCs) in turbine components with complex geometry.

5.Increased accuracy in detection, sizing, location and orientation of critical defects, regardless of their orientation. This requirement dictates multiple focused beams with the ability to change their focal depth and sweep angle.

6.Increased scanner reliability.

7.Decreased radiation exposure.

All the above requirements can be met by the following characteristics of phased array ultrasonic technology:

1.Flexibility: A single-phased array probe can cover a wide range of applications, unlike conventional ultrasonic probes.

2.Small probe dimensions: For some applications, limited access is a major issue, and one small phased array probe can provide the equivalent of multiple single-transducer probes.

3.Electronic setups: Setups are performed by simply loading a file and calibrating. Different parameter sets are easily accommodated by prepared files.

4.Speed: The phased array technology allows electronic scanning, which is typically an order of magnitude faster than equivalent conventional raster scanning.

5.Reliable defect detection: Phased arrays can detect defects with an increasedSNR, using focused beams. The probability of detection (POD) is increased due to angular beam deflection (S-scan).

6.Complex inspections: Phased arrays can be programmed to inspect geometrically complex components, such as automated welds or nozzles, with relative ease. Phased arrays can also be easily programmed to perform special scans, such as tandem, multiangle TOFD, multimode and zone discrimination.

7.Imaging: Phased arrays offer new and unique imaging such as S-scans, which permit easier interpretation and analysis.

9.6.2Meaning of Phased Array

An array transducer is simply one that contains a number of separate elements in a single housing and ‘phasing’ refers to how those elements are sequentially pulsed. A phased array system is normally based around a specialized ultrasonic transducer that contains many individual elements (typically from 16 to 256) that can be pulsed separately in a programmed pattern. These transducers may be used with various types of wedges, in a contact media, or in immersion testing. Their shape may be square, rectangular, or round, and test frequencies are most commonly in the range from 1 to 10 MHz. A typical multiplexer and array circuit is shown in Figure 9.14.

Phased array systems pulse and receive from multiple elements of an array. These elements are pulsed in such a way as to cause multiple beam components to continue with each other and form a single wavefront travelling in the desired direction. Similarly, the

Controller

To display console

Wave

front

Figure 9.14 Multiplexer arrangement for array transducers. Normally refracted, longitudinal wave shown (Bray and Stanley [21])

receiver function combines the input from multiple elements into a single presentation. Because phasing technology permits electronic beam shaping and steering, it is possible to generate a vast number of different ultrasonic beam profiles from a single probe assembly, and this beam steering can be dynamically programmed to create electronic scans. This enables the following capabilities:

1.Software control of beam angle, focal distance and beam spot size. These parameters can be dynamically scanned at each inspection point to optimize incident angle and SNR for each part geometry.

2.Multiple angle inspections can be performed with a single, small, multielement probe and wedge, offering either single, fixed angles or a scan through a range of angles.

3.These capabilities provide greater flexibility for the inspection of complex geometries and tests in which part geometry limits access.

4.Multiplexing across many elements allows motionless high-speed scans from a single transducer position. More than one scan may be performed from a single location with various inspection angles.

9.6.3Principle of Phased Array Ultrasonic Technology

Conventional ultrasonic testing uses monocrystal probes with divergent beams. In some cases, dual-element probes or monocrystals with focused lenses are used to reduce the dead zone and increase the detect resolution. In all cases, the ultrasonic field propagates along an acoustic axis with a single refracted angle. A single-angle scanning pattern has a limited detection and

Figure 9.15 Example of application of phased array ultrasonic technology on a complex geometry component. Left: monocrystal single-angle inspection requires multiangle scans and probe movement; right: linear array probe can sweep the focused beam through the appropriate region of the component without probe movement (Olympus NDT [23])

sizing capability for misoriented defects. Also, inspection problems become more difficult if the component has a complex geometry and a large thickness, and/or the probe carrier has limited scanning access. In order to solve the complex problem, a phased array multicrystal probe with focused beams activated by a dedicated piece of hardware will be required (Figure 9.15).

Take a mono-block crystal and cut it into many identical elements, each with a pitch much smaller than its length. Each small crystal or element can be considered a line source of cylindrical waves. The wavefronts of the new acoustic block will interfere, generating an overall wavefront with constructive and destructive interference regions. The small wavefronts can be time delayed and synchronized in phase and amplitude in such a way as to create a beam. This wavefront is based on constructive interference, and produces an ultrasonic focused beam with steering capability. A block-diagram of the delayed signals emitted and received from phased array equipment is shown in Figure 9.16.

The basic components required for a phased array ultrasonic scanning system is shown in Figure 9.17. The main feature of this type of system is the computer-controlled excitation (amplitude and delay) of individual elements in a multielement probe. The excitation of piezocomposite elements can generate beams with defined parameters, such as angle, focal distance and focal spot size through software.

To generate a beam in phase and with constructive interference, the multiple wavefronts must have the same global time-of-flight arrival at the interference point. This effect can only be achieved if the various active probe elements are pulsed at slightly different and coordinated times.

As shown in Figure 9.18, the echo from the desired focal point hits the various transducer elements with a computable time shift. The echo signals received at each transducer element are time-shifted before being summed together.