quantitative tomographic non destructive testing for ...

QUANTITATIVE TOMOGRAPHIC NON DESTRUCTIVE TESTING FOR FLAW SIZE MEASUREMENT AND LIFE TIME PREDICTION Uwe Ewert1, Bernhard Redmer1, Marc Kreutzbruck1 Andrey Bulavinov2, Udo Netzelmann2 1

2

BAM, Unter den Eichen 87, 12205 Berlin, Germany Fraunhofer-IZFP, Campus E 3 1, 66123 Saarbrücken, Germany [email protected]

Keywords: Radiography, ultrasound, thermography, computed tomography, dimensional measurement, flaw detection Abstract: Modern digital non destructive testing (NDT) techniques achieved an extraordinary improvement of measurement accuracy after transition from classical manual testing to computerized technology during the last decade. X-ray computed tomography (CT) is the most accurate NDT technology, but it was earlier restricted to in house laboratory applications. During the last 10 years a variety of CT systems has been installed in different production facilities for quantitative measurements of flaws, dimensions of inner structures and product integrity. Mobile and in-line CT is applied for casting inspection, aircraft and pipe weld analysis. A mobile Planar CT scanner enables the evaluation of pipe weld undercuts, lack of fusion and crack dimensions for in-service inspection. The quantitative measurement of flaw depth (ligament) and length for in-service inspection enables the prediction of life time for chemical plants and installations in power stations. The technique was qualified by open and blind trails on basis of the European Network of Inspection and Qualification. Tomographic inspections of aircraft components were carried out for evaluation of stringer integrity in carbon fibre reinforced components. New computerized ultrasound techniques on basis of a phased array design enable the 3D reconstruction of flaw dimension. The reconstruction technique is based on the synthetic aperture focusing technique (SAFT), which is similar to the unfiltered planar X-ray CT. The accuracy and presentation of flaw structure and dimension have been significantly improved. SAFT-B-Scan slices are combined in a data cube visualizing the flaws in 3D. Flaw sizing examples of steel castings and CFRP components of aircraft structures are presented. During the last years the thermographic testing was enhanced from simple heat imaging to flaw detection with lateral and depth measurement. New cameras provide more detailed information. Fourier techniques and phase analysis of active heat pulse induced time response measurements enable the 3D tomographic reconstruction of flaws and structure details with enhanced accuracy. New reconstruction results of pulse excited thermography for defect sizing are presented.

1.

INTRODUCTION

Traditional non destructive testing methods as e.g. radiography were applied over decades for visualisation of flaws in the material and flaw characterization. Modern strategies of critical engineering assessment and fitness for purpose require additionally the measurement of flaw sizes enabling the calculation of fracture mechanical data und life time prediction. This requires basically 3-dimentional (3D) analysis of flaws and its characterisation as e.g. crack or lack of fusion. The best way for an accurate evaluation of flaws is the inspection with Xray computed tomography (CT). X-ray CT was developed in the 70ties for small objects, which can be moved and positioned in a CT scanner. The development of modern digital detectors and specialized X-ray tubes as well as the availability of fast and affordable computers extended significantly the application range for industrial problems and the quality of CT results. New approaches for mobile application of CT have been developed and are applied for objects which cannot be brought to the testing device. The mobile technologies are

discussed in comparison to other 3D imaging technologies as UT phased array technique and pulsed dual band thermography. UT phased arrays have revolutionised the UT inspection and enable presently the visualisation of flaw distributions in test objects. Synthetic aperture focussing techniques enable the reconstruction of 3D images in analogy to the computed tomography. Different approaches have been developed for fast and accurate measurements. The pulsed phased array technique has been optimized for fast and accurate 3D reconstruction, sizing and characterization of flaws. Thermography was developed from a measuring tool for temperature distributions at surfaces to a visualisation tool for internal 3D distributions in test objects by application of the thermal pulse technique and detectors which enable energy and time resolved imaging. 2.

MOBILE COMPUTED TOMOGRAPHY

One of the key tasks of NDT and especially of in-service inspection (e.g. welds) is the detection of planar defects and its sizing. The depth, length and kind of the indications are essential features for the decision about acceptance or rejection and repair. This knowledge can also support life time prediction. If the maintenance of pipes and other industrial products is performed as result of NDT, longer periods between maintenance cycles can be accepted and the life time of the component can be extended. Mobile X-ray CT systems were developed for analytical inspection of different objects, which cannot be moved or brought into a laboratory. This is important for industries with high safety requirements as e.g. nuclear power industry and aerospace industry. The increase of the accuracy of NDT measurements should also enable an increase of the time between inspections or repair cycles. 2.1

Mobile CT in nuclear power industry

Fig. 1: Principle of Tomographic Computer Aided Radiometry (TomoCAR) for mobile inspection.

The mechanised mobile tomographic system “TomoCAR” (Tomographic Computer Aided Radiometry) was developed originally for inspection of circumferential welded seams of pipes [1-3]. It consists of the manipulator based position control of an X-ray tube in front of the region to inspect and a Digital Detector Array (DDA) behind it (fig. 1). TomoCAR, was developed and qualified according the ENIQ (European Network of Inspection and Qualification) standard during the last years [3]. The system was designed to scan

the test sample (e.g. girth seams of pipelines) using a digital detector (e.g. line camera, flat panel detector) and an X-ray tube under a computer controlled geometry. The developed planar TomoCAR technique is based on a continuously shift of an X-ray tube [5] parallel to the sample in a radiation angle range of about ±45° to the detector normal. During the movement of the X-ray tube usually more than 400 two dimensional projections are acquired. Each single scan represents a defined angle of incidence of the X-ray beam in relation to the detector and sample. The detector is not shifted during the scan of the X-ray tube. Fig. 2 shows a TomoCAR device, mounted on a pipe in a power station. Certification by European uropean  Ne   twork of IInspection and Qualification. Q   Metallography

8,5

Planar Tomography

8,0

Fig. 2: TomoCAR mounted on a pipe for digital radiography and CT to measure non destructive cross sections in comparison to destructive micrographic inspection.

The tomographic reconstruction allows the three-dimensional (3D) representation of the material structure and included flaws. This is equivalent to a metallographic cross sectioning (see fig. 2, micrography). TomoCAR is applied for sizing of volumetric and planar defects. The physical pixel size of the selected detector [5] amounts to 0.1 x 0.1 mm². It allows the reliable detection of planar defects with openings larger than 25 µm by subpixel resolution due to the achieved high signal-to-noise ratio. TomoCAR has been used for sizing of planar flaws several times in different nuclear power stations in Germany and Switzerland. 2.2

Gantry based CT for aircraft inspection

The TomoCAR design was modified for in situ inspection of large aircraft components under production conditions [4]. A gantry gate based planar tomograph was constructed and tested for inspection of the integrity of large flat panel components (fig. 3). The TomoCAR geometry was modified and equipped with a specially designed X-ray tube of COMET and a large DDA of PerkinElmer. CFRP panels (Carbon Fibre Reinforced Polymers) of aircrafts of up to 3 x 9 m size were inspected. The integrity of imbedded stringers was successfully tested. No alternative NDT method with similar detection rate has been found up to now. The technology was tested and qualified by the Airbus Company. Different trials were performed to prove the finding rate of cracks in embedded stringers. The test sample shown in fig. 4 was a spherical panel of a vertical tail component of a plane. At both ends of the sample, 87 positions were defined and measured (marked in yellow in fig. 3). The total length of all measured areas amounted to about 14 m. Fig. 3 shows the result of two planar reconstructions of embedded stringers. The black/white frame lines in fig. 3a, b mark the geometrical shape of the T-stringer in the embedding CFRP environment. The white stringer tip was outside the CFRP panel and the grey one inside, that means embedded. The two black indications in Fig. 3a show the adhesive connection of the

stringer to the basic CFRP panel. The stringer foot is visible as “roughness”, indicating a horizontal line with white end points. For qualification of the TomoCAR method, several embedded stringer components were manufactured with high stress. These test samples showed stress cracking which had to be found. Fig. 3b shows such a cracked embedded stringer (60% embedded). The crack shape and length can clearly be seen and the dimension was measured. The qualification of the gantry based TomoCAR technique was successfully performed by comparison of the tomograms to cross sections after destructive micrography of test samples. The probability of visibility of cracks was > 90% at a confidence level of 95%.

a) Fig. 3: Moving Gantry Bridge. a) The detector and the X-Ray tube were positioned on the left and right side of the sample. The X-Ray tube was moved on the linear axis, which was oriented perpendicular to the stringer. b) The spherical shell of the vertical tail was located between the X-ray tube and the detector. The size of the panel amounted to 9 m x 3 m. The gantry bridge could be moved parallel to the sample.

a) 

b)

b)

Fig. 4: Result of the cross section reconstruction of a part of stringer: a) with “good” indications and b) with “bad” indications. The black/white lines mark the real geometry of the partly imbedded stringer and the embedding CFRP material.

3.

ULTRASONIC INSPECTION WITH PHASED ARRAYS

The radiographic inspection provides the most accurate results but it also has some disadvantages. The radiographic and computed tomographic inspection requires access from both

sides of the object to inspect. Many attempts have been undertaken to solve this problem by application of X-ray back scatter techniques. Most back scatter devices are not sensitive to fine planar defects and provide less image quality than the transmission technique. Another disadvantage of RT and CT is the requirement to protect operators against ionizing radiation. Therefore, the UT technique was improved during the last years, enabling 3D inspection of flaws with single sided access. Efficient UT imaging techniques are performed with phased array UT systems. Different UT transmitter and receiver heads are implemented in a sensor line or a 2D array. Applications with phased arrays can be performed in different ways. The UT beam can be scanned in 1 or 2 space directions to acquire the 2D or 3D geometry of a flaw in a material. This corresponds to the conventional UT approach, when one or two UT heads were manipulated on the object surface and echo sequences were taken and reconstructed to image information. The UT sequences are acquired and can be reconstructed with the SAFT (synthetic aperture focussing technique) method. 3.1 Conventional and Sampling Phased Array Technique The so called “Sampling Phased Array” technique which is different to the conventional phased array approach, became very popular in the last years making use of the propagation of elementary waves (Huygens principle) generated by individual elements of the sensor array to reconstruct the composite phased array signal by high-speed numerical computation for any arbitrary, individual angle or physically possible focus depth. As the near field of each individual transducer in the array is that of a ‘quasi’ point source, transducer-near inspection regions are better to inspect than in the conventional case, where the near field is determined by the full aperture of the transducer array. Furthermore, a fast 3D visualization of inspected objects can be performed by combining - slice by slice of 2D calculated SAFT-Bimages to a 3D data-cube.

2xL

2xL L

Conventional  Phased Array 

L

The physical sound beam, steered by an electronic transmitter/receiver delay control unit, allows the Sampling Phased Array technique (SPA) reconstruction of sector image by synthetic composition of elementary wavelets [6, 7]. Thus after a transmitting pulse a sector image containing all angles of incidence can be reconstructed, e.g. by using the Kirchhoff Migration algorithm [8]. Thereby, synthetic focusing occurs for every image point (Fig. 5).

Sampling Phased  Array 

Fig. 5: Principles of Conventional Phased Array (left) and Sampling Phased Array (right).

The reconstruction principle can be described as follows (Fig. 6). For all probe positions (x), the complete High Frequency signals are acquired for tomographic reconstruction of the inspection volume. Because of the large beam spread, the overlapped echo signals from different reflectors are simultaneously received and saved. The captured volume represents one plane perpendicular to the surface of the object to inspect. Propagation times from each probe position to each volume pixel on the plane and back are calculated. Similar to the SAFT algorithm, the (Fig. 6) cross section or volume of the component is divided into pixels. Selected point defects represent the elements to detect (see Fig. 6.1). For each such point element, the travel time to the individual transducer elements is calculated (Fig. 6.2) and the time-related amplitude value is assigned. Fig. 6.3 and 6.4 illustrate the

image reconstruction principle for a single point element as a function of element location with a hyperbolic distribution of the travel time. x1 

x2 

x3 

z1 

z1

z2 

z2

z3 

z3

Z

Point Reflectors in the  Reconstruction Volume  x1 

x2 

x3

t

X

z1

z2 

z2

z3 

z3

Superposition Principle 

Z

x3

X

x3

X

Data Hyperbola

x1

z1 

Z

x2

x1

X 

x2

Result of reconstruction

Fig. 6: Reconstruction principle of Sampling Phased Array technique.

3.2. Industrial application of Sampling Phased Arrays with 3D imaging 3.2.1 In situ casting inspection by a mobile SPA unit

Fig. 7: SPA testing of a casting object

Material flaws in castings posses often pure reflectivity. Thus typically large defects like cavities or inclusions cannot be correctly evaluated by standard UT techniques. Fig. 7, represents the ultrasonic inspection results by the SPA technique on a massive casting object. The real defect size measured by the SPA method and approved by metallographic analysis was about 80 x 40 x 30 mm. The three-dimensional tomographic reconstruction of the inspected object allows the quantitative evaluation of the inspection results. The mobile inspection system consists of a two-axes manipulator with a phased array transducer and UT electronics mounted on it. The evaluation station (master-PC) is capable of real time representation of the 3D image while scanning is performed. A very fast analysis of material flaws can be provided to the customer as a mobile service. 3.2.2 Testing of CRFP components The ultrasonic testing of CRFP components has a number of distinct challenges when compared with other applications. The practical use of ultrasonic testing is strongly correlated with the defects prevalent for this type of material, especially lamination, porosity, plastic or paper films, undulations. Due to lower wall thicknesses and in-plane orientation of material defects, normal probes are often used for ultrasonic testing of CFRP-components. But for complex geometries like Corner joints and T-joint inspection, Phased array transducers are applied. The Phased Array testing of light weight components is the state-of-the-art in the aircraft industry since several years. Large inspection area of the components can be optimally inspected by wide phased array transducers with more elements where, the so called linearscan function option is applied for fast covering of the inspection volume. SPA technique permits an increase in the inspection speed and improves the flaw detectability due to the synthetic focusing at every point in the image [9]. Thus the defects in all locations (incl. first and last CFK layer) can be reliably detected for planar and curved geometries (corner joints) followed by the three-dimensional representation of the image in real time. Fig. 8 represents inspection results of a test specimen with artificial defects by means of SPA. Scanner path PA probe

3D volume

Fig. 8: Data acquisition and reconstruction principle for testing of CRFP components with SPA

3.3 SAFT using Phased Arrays Alternatively to the Sampled Phased Array Approach (SPA) the SAFT was applied for improved e2D and 3D data acquisition and image interpretation. SAFT generally requires transducers, which provide a sufficiently large sound beam divergence, similar to the SPAmethod, where the elements of the phased array are excited sequentially and other elements

act as receivers. In the following application the PA generates a large divergence by subsequently using different swivel angles. The phased array probe as a whole has a swivel angle range which is determined by the divergence of the sound beam for each individual transducer element. It is thus capable of sending and receiving sound in an equally wide angular range as a single small probe. In doing so, one can obtain the additional benefit of a high SNR, as the sound field of a single angle gets the energy of all transducer elements. The sound beam is swivelled at each measuring point, a HF A-scan is taken for each swivel angle and all A-scan data are processed by the SAFT reconstruction. This procedure was proposed and examples were demonstrated a decade ago [10]. The instrument and computer technology used at that time prevented the technology from being embraced by the industry. The situation is different today [11] because of the tremendous increase of calculation power. The amount of raw data is increased by the number of swivel angles which are needed in order to cover the desired aperture with the given sound beam width. Even if the respective sound beam width is taken into account in the SAFT algorithm, the calculation time for SAFT reconstructions is not extended. 850

3-4

y [mm]

855 723,5

860

1-2

865

205

210

215 x [mm]

220

225

Fig. 9. Left: Sketch of a turbine disk segment, with a diameter of 800 mm, containing FBH (1-2, 3-4) with a diameter of 1mm. Middle: A-scan of a FBH with 1mm in dia. when using a 3 MHz PAtransducer. Average angle of incidence 12.5°, scanning path: 0°-30° with a sampling distance of 0.2° (3mm). Right: SAFT-scan of the 1mm-FBH.

Fig. 9 shows the detection of a 1mm sized flat-bottom-hole in a segment of a turbine disk in which the propagating paths is about 900 mm. Due to the good steel quality frequencies of up to 3 MHz were used. The FBH appears about 20 mm in front of the back wall echo and can be detected with a maximum SNR of about 7 dB (an example is shown in Fig. 9 middle). For the SAFT the same PA-transducer was used and the data along a scanning path of 30° with a sampling distance of about 0.2°, corresponding to a point-to-point separation of about 3 mm, were reconstructed. As a result the FBH with an increased SNR of about 17 dB could be detected. In a second example the results of a test turbine shaft (Siemens test specimen 4D290) with a diameter of up to 1600 mm diameter and a weight of several 10 tons are present. The measurement was performed with a frequency of about 4 MHz containing several small defects. Figure 10, top shows the SAFT scans, in which all data of the 360° measurement were used for the tomographic SAFT reconstruction. Here, the reflections caused by the change in diameter and two 5 mm-sized side drilled holes were observed. When magnifying the core regime of the shaft, a number of indications (Fig. 10, bottom left), similar to a cluster of stars were observed. A further zoom out reveals the defects as 1 mm – 4 mm large indications in a field of an increased noise level, which can not be resolved when inspecting single A-scans. This result shows that SAFT is a powerful visualisation tool for detection of small defects even in very thick steel parts. Here the SAFT algorithm increases both the signal-to-noise ratio and the spatial resolution as well. It is mentioned that UT-SAFT sizing is generally limited in the same way like conventional UT inspection with respect to the proper sizing of defects. In case of smaller reflectors with sizes significantly below the defect dimensions, amplitude criteria like the DGS method should be

used for the size estimation. However, SAFT makes the size estimation easier, as the spatial resolution can be distinctly increased.

200 mm

10 mm

Fig.10. Top: Sketch of a turbine shaft, with a diameter of up to 1600 mm, containing 2 side drilled holes and a large number of small defects within the core regime. Right: SAFT-scan of the complete shaft. Bottom: SAFT-scan of the core regime showing a magnification with an area of 400 mm x 400 mm (left) and 100 mm x 100 mm (right).

4.

PULSED THERMOGRAPHY

Pulsed and dual energy thermography is a new NDT method with increasing potential. It allows the 3D reconstruction of structures in materials in a short time. The technology is under development for new application fields and is discussed as an emerging new NDT technique. The inversion of thermal diffusion processes is generally an ill-posed problem on recovering the multi-dimensional shape of hidden structures. Recent work was performed at the tomographic reconstruction of the defect shape in a test sample with known thermal properties after pulsed thermal surface excitation [12]. The developed inversion algorithm consists of a defect shape correction unit and a simulation unit. A first approximation of the defect shape is based on a 1D analysis of the time dependent information for each pixel. The simulation unit models the heat conduction process in the inspected sample in two or three dimensions based on an approximation. Using the result of the forward calculation, the defect shape correction unit improves the initial approximation. In several iterations the defect shape will be obtained.

Fig. 11: Top: Photograph of a steel plate with a simulated corrosion profile at the back side. Bottom: Experimental thermographic reconstruction of the back side shape when applying the algorithm in several iterations a) to d). The black lines represent the true shape, the dashed blue lines represent the reconstructed profile.

The algorithm was tested using experimental data obtained on a plate-shaped steel sample with a wall thickness profile (Fig. 11). The specimen was heated by a flash pulse from the top side. From the same side, thermographic cooling sequences were recorded by an infrared camera. Fig. 9 shows, how a low number of iterations leads to a quite accurate defect shape reconstruction. By the same approach, the shape of inclined cracks could be obtained within the principal limits given by thermal diffusion physics. Determination of defect depth or coating thickness in thermographic testing requires the knowledge of thermal material properties. When working on infrared transparent materials like polymers or ceramics, the knowledge of optical properties in the thermal infrared range is required in addition. New dual energy infrared cameras are now available that give access to pixel- and time synchronized information in two spectral bands. Defect depth or coating thickness are usually determined by measuring the time after pulsed excitation, where the heat flow is notably blocked by the defect or disturbed by the substrate interface. The effect of the partial optical transparency in the infrared region is demonstrated [13] on a PVC polymer sample with a notch with 4 mm width and 1 mm ligament on its back-side, which was investigated by light flash pulsed thermography (Fig. 12). A dual-band IR camera sensitive in the mid-wave infrared (MWIR) at 4.4-5.2 µm and in the long wave infrared (LWIR) at 7.8-8.8 µm wavelength was employed for this measurement.

radiation difference (a. u.)

10000

1000 MWIR reference LWIR over notch LWIR reference MWIR over notch

100 0.01

tMWIR tLWIR 1

0.1

10

time in s

Fig. 12: Radiation decay curves in PVC after flash heating in two spectral ranges in the MWIR and the LWIR. For both spectral ranges, the times of beginning defect contrast tMWIR and tLWIR are indicated. Inset: Sample view from the back-side

The radiation difference due to heating is shown in Fig. 12 as a function of time after excitation. The result shows, that the cooling slopes are dependent on the spectral range, which is due to the different translucence of the polymer in the two spectral ranges. Moreover, the point in time, where the cooling curve over the defect deviates from that of the cooling curve in a defect-free region of the sample is different in the two spectral ranges. This would result in a serious difference in the measured defect depth, if the spectral effect was not considered. It is known that during ageing or thermal cycling some materials change their infrared optical properties. The additional spectral information offered by the camera will allow to control such effects and will lead to more reliable defect depth measurements. 5.

CONCLUSION

Mobile Tomographic Computer Aided Radiometry (TomoCAR) is based on the mechanical position control of an X-ray tube in front of an object and the application of a Digital Detector Array (DDA) behind it. Several hundred radiometric projections in small angle steps are acquired during the controlled movement of the X-ray tube. A German pilot study for qualification of TomoCAR for nuclear power industry was successfully performed on the basis of the ENIQ guidelines. This was the precondition for successful applications of TomoCAR in different nuclear power stations in Germany and Switzerland for sizing and characterization of inhomogeneities. The basic design and working principle of the “TomoCAR” system was modified for the construction of a gantry bridge based planar CT system for inspection of large area CFRP panels of aircrafts. Cracked stringer shafts could be distinguished from properly manufactured ones in the reconstructed CT images. No cracks were found in the properly manufactured panels. A probability of crack visibility > 90 % could be determined at a confidence level of 95% in the CT images. Phased Array (PA) UT systems provide 3D-data, which can be reconstructed from conventional PA scans of position and beam angle in conjunction with SAFT, or sampled phased arrays (SPA). Fast inspection systems can be used for off-line analysis of safety-relevant components as well as for automated in-line testing in the production process. Advantages of the SPA technique due to the synthetic aperture principle were exploited for inspection of volumetric flaws, occurring over a large scanning area in castings and thick walled steel ob-

jects. The inspection was carried out with improved spatial resolution and signal-to-noise. The modified technique was also used for 3D inspection of fibre reinforced polymer components in aircraft industry. The features of the Sampling Phased Array technique, like fast three-dimensional imaging of material flaws in the inspection volume, is a great asset for automated high speed testing systems. It has high potential for testing of components with different shapes e.g. with complex geometries, where the indications have to be represented in the correct position with higher spatial resolution for appropriate defect sizing. The technique was used for different applications successfully, which allow only single sided access. Pulsed thermography, applied within different spectral ranges, was developed during the last decade for 3D inspection of metals, polymers, ceramics and building structures. New reconstruction techniques enable a better 3D geometry and flaw characterization. The reconstruction speed is still slow, but will increase with progress in computer design. New developments in infrared detector technology give access to spectral information which allows fast material characterisation near the surface and more robust defect characterization. The correct reconstruction of the shape of a corroded plate is demonstrated. REFERENCES [1]

Ewert, U. et.al.: „Digital Laminography“, Materialforschung 37 (1995), pp.218-222.

[2]

Redmer, B. et. al.: „Mechanised weld inspection by tomographic computer-aided radiometry (TomoCAR)“, Insight Vol. 44 (2002) No.9, pp.564-567.

[3]

REDMER, B. EWERT, U., NEUNDORF, B., JAKOB, M., “Tomographic 3D-Radiometry for the Visualisation and Measurement of the Defects of Girth Seams”, 9th ECNDT, 25. – 29. Sept. 2006, Berlin, Germany, http://www.ndt.net/article/ecndt2006/doc/We.3.2.3.pdf.

[4]

EWERT, U., REDMER, B., RÄDEL, Ch., OSTERLOH, K., SCHNARS, U., HENRICH, R., SCHIMMELMANN, O., BAVENDIEK, K., JAHN, M., Mobile X-Ray Inspection of Light Weight Materials, International Symposium on NDT in Aerospace December 3-5, 2008, Fürth, Germany, http://www.ndt.net/article/aero2008/aero08_ewert.pdf.

[5]

RTW RÖNTGEN - TECHNIK DR. WARRIKHOFF KG, http://www.rtwxray.de/index.php?id=oemdesigns , Oy AJAT Ltd, Technical description of DIC100T, http://www.ajat.fi/products_dic.php.

[6]

Kröning, M., Hentschel, D., von Bernus, L., Bulavinov, A., Reddy, K. M.; Verfahren zur zerstörungsfreien Untersuchung eines Prüfkörpers mittels Ultraschall, Deutsches Patent DE 10 2004 059 856 B4 2006.09.14.

[7]

Bulavinov, A.: Der getaktete Gruppenstrahler. Universität des Saarlandes, Saarbrücken, 2005 Dissertation.

[8]

Claerbout, J., F., Cecil and Green, I., EARTH SOUNDINGS ANALYSIS: Processing versus Inversion, March 23, 2004.

[9]

Bulavinov, A., Hanke, R., Hegemann, J., Kröning, M., Oster, R., Pudovikov, S., Reddy, K.M., Ramanan, S. V.; Application of Sampling Phased Array Technique for Ultrasonic Inspection of CFRP Components, International Symposium on NDT in Aerospace, December 3rd to 5th, 2008 Fürth, Germany.

[10]

T. Hauser, H.-J. Montag, R. Boehm, U. Voelz „Vergleich von Rekonstruktionsverfahren auf der Basis von Gruppenstrahler-Ultraschalldaten“ (Comparison of reconstruction procedures based on phased array ultrasound data), annual meeting of DGZfP, Bamberg 1998

[11]

J. Pitkänen, P. Oy: SAFT – Is it a Tool for Improved Sizing in Ultrasonic Testing, proceedings of the ECNDT 2006, Berlin Lugin, S. ; Netzelmann, U.: A Defect Shape Reconstruction Algorithm for Pulsed Thermography. In: NDT&E International. 40 (2007), pp. 220-228.

[12]

[13]

Abuhamad, M. ; Netzelmann, U.: Dual-band active thermography on infrared transparent materials. In QIRT Journal 2010, to be published.