ULTRASONIC NDE OF CONCRETE

Martin Schickert* Institute for Materials Research and Testing (MFPA Weimar) at the Bauhaus-University Weimar, Germany, 99423 Abstract In this contribution an overview of ultrasonic non-destructive evaluation (NDE) of concrete is given. The intention is to provide insight into the specific features of concrete testing, and to build a picture of this rapidly developing field by presenting examples of research results and applications. After introducing concrete as a construction material and inhomogeneous propagation medium, common inspection tasks at civil structures are identified. Characteristics of ultrasonic wave propagation in concrete are explained together with measurement and modeling methods. A short presentation of measurement systems precedes the description of research and applications examples. This subjective selection comprises measurement and signal processing techniques including imaging. The list encompasses well-tried methods as well as topics of current research. Finally, the current state of development is discussed in view of demanded testing objectives. I. INTRODUCTION Concrete is a material used for buildings, bridges and the infrastructure in general. Its composition is adaptable to many tasks, and concrete elements can be made in a great variability of shapes. This flexibility has made concrete structures popular. Construction defects, wear, and aging can deteriorate concrete making inspection and quality control necessary. In addition to conventional destructive probing, non-destructive techniques have been developed with the intention of easy, expressive, and reliable assessment of concrete structures. For this purpose, ultrasonic techniques have been used for more than four decades [1]. In the beginning, the main objective was the measurement of ultrasonic bulk velocity and attenuation and their empirical correlation to material properties such as concrete strength. During the last 15 years adopting methods developed in other fields of ultrasonics enhanced possibilities. New techniques emerged for the detection and localization of structural components and defects that gradually find their way into practice. This contribution is intended to describe concrete as a subject of ultrasonic testing, and to summarize prior and actual research efforts towards NDE applications. An earlier survey has been conducted in [2]. Covering the years 1975-1983, an annotated bibliography of non-destructive evaluation methods including ultrasonic testing is contained in [3]. II. CONCRETE AND CONCRETE STRUCTURES Hardened concrete consists of three components: aggregate, cement, and pores (Fig. 1). Aggregate takes up 70 % of the volume, and consists of sand and gravel stones in most cases. For a certain recipe, aggregate of different size fractions is mixed according to a grading curve. Depending on the application, the maximum aggregate size is chosen to be 8, 16, or 32 mm. These values are standardized in each country, the values given are for Germany. The aggregate is bound by a cement matrix occupying 15 Vol-%. Within the matrix and to a less extend the aggregate, air pores are contained. Porosity has various origins, and, as a result, connected and disconnected pores are distributed in a wide range of sizes from 1 nm to 10 mm with the largest fraction in the micrometer region. Porosity totals to 15 % of the concrete volume. The given volume fractions are for an average concrete and may vary.
Aggregate Cement matrix Pores

Fig. 1. Composition of concrete

0-7803-7582-3/02/$17.00 (c) 2002 IEEE

2002 IEEE ULTRASONICS SYMPOSIUM-739

Authorized licensed use limited to: IEEE Xplore. Downloaded on December 7, 2008 at 06:50 from IEEE Xplore. Restrictions apply.

Concrete accepts large compressive loads but is sensitive to tensile traction. To improve its behavior under tensile stress, steel bars are cast in. Two construction principles are used: reinforcement and posttensioned tendons. Reinforcement consists of single steel bars of 8 to 28 mm diameter. It is laid out in one or more layers of parallel bars or meshes 20 to 60 mm below each surface of the concrete. The bars take part of the tensile load. Pre-tensioned tendons are ropes of several steel bars contained in a tendon duct of 30 to 100 mm diameter (Fig. 2). The gaps are filled with injected mortar. After hardening of the concrete, stress is applied to the tendons. As a result, the concrete is pre-loaded with compressive strength, and more load can be charged until a tensile state is reached. Pre-tensioned tendons are placed in the inner part of concrete elements.
Concrete Duct

Tendons Void Injection mortar

Crack: The depth of surface-opening cracks is important to assess the structural condition. Honeycombing: The appearance of voids as a result of a lack of cement paste is named honeycombing. It often occurs close to edges and between reinforcement. Of interest are position and size. Voids in tendons: Voids in tendon ducts involve the risk of rusting and eventual failure of tendons (Fig. 2). Micro-cracking: Certain chemical processes can cause areas of micro-cracking thus reducing concrete strength. Layers: If a concrete element is set-up of different layers or a repair layer is applied, thickness and boundary condition may need to be determined. Compressive strength: The achieved strength is among the most important concrete properties. Elastic moduli: For the development of concrete recipes, elastic moduli are frequently evaluated. Hardening: Construction time can be shortened if the hydration process can be monitored.
Position Thickness Crack Void Boundary condition Layer

Thickness

Fig. 2. Cross-section of a post-tensioned tendon

III. COMMON EVALUATION TASKS A typical concrete slab, e.g. in a bridge deck, consists of concrete with reinforcement close to the surfaces and embedded tendon ducts (Fig. 3). Using this example, a number of common inspection and evaluation tasks can be identified:

Corrosion Honeycombing Micro-cracking

Fig. 3. Common inspection tasks at a concrete slab

Thickness: Thickness measurements are a frequent demand in cases where the back-side of concrete is buried in sand or rock, or is not easily accessible. Position: The exact lateral and depth position of tendon ducts and reinforcement may differ from the construction plan and needs to be determined. Corrosion: Larger areas of corrosion of reinforcement and corrosion in tendons weaken the construction.

Any technique considered for the testing of concrete structures has to fulfill requirements specific for applications in civil engineering. Most important, results need to be clear and unambiguous. Evaluation results can cause far-reaching decisions, thus customers require statements such as defect/no defect, position or thickness in centimeters, concrete element is safe/ not safe. In order to achieve this goal, it can be necessary to apply more than one nondestructive testing technique, possibly backed up by destructive probing. Then, methods requiring onesided access only are advantageous in cases where the back-side is not easily accessible. Two-dimensional maps or images of the results ease interpretation.

2002 IEEE ULTRASONICS SYMPOSIUM-740

Authorized licensed use limited to: IEEE Xplore. Downloaded on December 7, 2008 at 06:50 from IEEE Xplore. Restrictions apply.

Additionally, result interpretation should be made not only from a non-destructive, but also from a civil engineering point of view. Last not least, application of the techniques should be fast, with small effort, at reasonable costs, and should not interfere with construction work or traffic. In general an application-oriented approach, not a method-oriented approach is demanded in NDE of concrete. IV. ULTRASONIC PROPAGATION IN CONCRETE The three components aggregate, cement matrix, and pores constitute concrete as a heterogeneous material. In order to estimate consequences for the propagation of ultrasonic waves, Table I contains typical acoustic impedance values of the three components, and (plane wave) reflection factors relative to the cement matrix. These values indicate that concrete is a strongly scattering medium.
TABLE I ACOUSTIC IMPEDANCES AND REFLECTION FACTORS RELATIVE TO THE CEMENT MATRIX FOR LONGITUDINAL WAVES

TABLE II TESTING PARAMETERS: MAXIMUM AND TYPICAL RANGES FOR LONGITUDINAL WAVES Parameter Diagnosis range Frequency range Sound velocity Wave length Divergence angle Maximum Typical

0 1000 mm 50 500 mm 20 500 kHz 50 200 kHz 3500 5000 m/s 4000 4500 m/s 200 8 mm 90 20 mm 20 180 40 100

Component Cement matrix Aggregate Air pores

ZL 7 MRayl 17 MRayl 0.4 MRayl

RL 0.4 0.9

As a consequence, scattering is the most prominent cause for attenuation of an incident ultrasonic pulse on its way through the concrete. Assuming the exponential attenuation model in (2), the attenuation coefficient has been empirically determined to = 5.6 dB/(MHz cm) (1)

(evaluating the results from [4], see below). This value is about five to 10 times larger than the one characterizing biological soft tissues [5]. Incoherent scattering at the concrete structure of aggregate and pores sums up to so called structural noise in the received signals. The strong attenuation of ultrasound in concrete has consequences for the ultrasonic testing procedure. Table II summarizes some test parameters in their typical and maximum ranges. The values are given for longitudinal waves and may be exceeded in special cases.

Both the need to penetrate concrete to large depths and strong attenuation restrict the frequency range to the lower bound of ultrasonic testing. In combination with typical sound velocities, wavelengths of a couple of centimeters follow, thus limiting resolution. Scattering at pores occurs mainly in the Rayleigh regime, scattering at the aggregate extends from the Rayleigh to the Mie region. Since wavelengths and transducer diameters are of the same order, broad divergence angles result. Concrete as a solid allows propagation of longitudinal, transversal, and Rayleigh waves. Mode conversion occurs at all boundaries at objects as well as at aggregate and pores. Because of the low relation between transducer diameter and wavelength, generation of Rayleigh waves can be prominent. As received signals contain a superposition of excited, reflected, and mode converted waves, signal identification especially in single A-scans may be difficult. Fig. 4 shows examples of received signals in transmission and pulse-echo-mode. Please note that the plots have different time scales. In transmission mode [Fig. 4(a)], the signal first received corresponds to the excited pulse, followed by structural noise (coda). Structural noise is also present in received pulse-echo signals at all time instances [Fig. 4(b)]. As in this example, indications of object reflections are often hard to detect from single time signals. In Fig. 4(b), noise shape smoothens in the course of the signal. This effect is due to the behavior of concrete as an acoustic low-pass filter, based on frequency-dependent sound attenuation [4], and, to a minor extend, on dispersion [6]. Fig. 5 depicts the filtering effect for transmission signals in both the time and the frequency domain. In either case, a reference signal measured at a homogenous material is compared to signals received after sound paths of 100 and 500 mm, respectively, in concrete. The coherent parts were extracted by time-

2002 IEEE ULTRASONICS SYMPOSIUM-741

Authorized licensed use limited to: IEEE Xplore. Downloaded on December 7, 2008 at 06:50 from IEEE Xplore. Restrictions apply.

windowing. With increasing sound path, signals broaden in the time domain, and their spectra are shifted to lower frequencies in the frequency domain. As a consequence, broadband transducers are employed to balance between resolution and penetration depth. These graphs also show that in ultrasonic concrete testing all signal parameters become dependent of depth and, therefore, time.

models do not seem to be able to handle such arrangements.

Fig. 5. Effect of frequency-dependent sound attenuation: (a) time domain, (b) frequency domain

Fig. 4. Examples of received signals: (a) transmission mode, (b) pulse-echo mode

V. UNDERSTANDING PROPAGATION Understanding the propagation of ultrasonic waves in concrete is essential for the successful development of measurement systems and signal processing techniques. Important aspects of this task are first to investigate the propagation of ultrasonic waves depending on concrete parameters such as aggregate size, porosity distribution, and strength. Then, the interaction of ultrasonic waves with flaws and objects such as reinforcement and tendon ducts needs to be clarified. Keeping the number of influential material parameters in mind, this turns out to be an extensive task. Analytical modeling is usually best suited to make dependencies obvious. But considering the large number of densely packed, strong scatterers, present

Numerical models exist in a wide variety of, e.g., ray and grid-based models. During the past years, particular progress has been achieved using the Elastodynamic Finite Integration Technique (EFIT) [7], [8], [9]. Fig. 6 shows a certain time instant of the propagation of longitudinal waves, transversal waves, and structural noise, excited by the transducer at the bottom.

Fig. 6. Numerical modeling of wave propagation using the Elastodynamic Finite Integration Technique (EFIT) [10]

2002 IEEE ULTRASONICS SYMPOSIUM-742

Authorized licensed use limited to: IEEE Xplore. Downloaded on December 7, 2008 at 06:50 from IEEE Xplore. Restrictions apply.

Measurements at test specimen have the advantage that true conditions are treated without restrictions. On the other hand, every situation has to be built separately. This is a problem for concrete, which is hardly reproducible even for identical recipes, and which takes weeks to harden. Also, size and weight of the specimen grow easily to values that are hard to handle if boundary reflections are to be avoided. Fig. 7 shows three concrete blocks as an example for small test specimens. Their dimensions are 300 x 300 x 300 mm, and their masses are about 60 kg.

Measurements at concrete specimens whose properties are varied systematically may lead to models of ultrasonic propagation that can be classified as empirical modeling. A model for the frequencydependent sound attenuation of the coherent signal in concrete was proposed in [4]. It uses an exponential approach for the magnitude of the acoustical transfer function |H| of the form | H | = exp[(s, g, p) f ] , (2)

where f is the frequency and the attenuation coefficient dependent on sound path s, maximum aggregate size g, and porosity p. Employing linear correlation with measurements and setting some correlation coefficients constant, the final model reads | H | = exp[(As + Bg + C) f ] . (3)

Fig. 7. Concrete test specimens with varied maximum aggregate size and lateral drillings

Larger specimens containing tendons and reinforcement can exceed a volume of 2 m and a weight of 4.5 t [11]. For field trials, structures to be demolished are particularly attractive. As an example, comparative studies of different non-destructive techniques were carried out at a bridge girder [12]. In such cases results can be compared to destructive examinations. A two-dimensional model concrete with dimensions of 300 x 100 x 400 mm uses glass bars of 3 to 14 mm diameter as a substitute for aggregate. Fig. 8 shows the specimen prior to filling with mortar [13].

The values of the coefficients are determined to A = 0.065 s/mm, B = 0.17 s/mm, and C = 4.5 s. Model and measurements are compared in Fig. 9 for a concrete of 16 mm maximum aggregate size. This model is preliminary in that porosity is not included (see also [6]), and geometric dispersion is neglected.

Fig. 9. Empirical propagation model: comparison of model (solid lines) and measurements (dotted), parameter is the sound path s

VI. MEASUREMENT SYSTEMS Ultrasonic measurement systems for concrete applications can be divided in two groups: Transmission and pulse-echo systems. Transmission systems are used for determination of ultrasonic pulse velocity and attenuation. During the last decades, a variety of such systems have been developed. Two transducers, sender and receiver, are

Fig. 8. Two-dimensional model concrete specimen using glass bars as aggregate [13]

2002 IEEE ULTRASONICS SYMPOSIUM-743

Authorized licensed use limited to: IEEE Xplore. Downloaded on December 7, 2008 at 06:50 from IEEE Xplore. Restrictions apply.

to be connected to the main unit (Fig. 10). Usually low frequency (20 to 100 kHz) transducers with low damping are employed offering high-energy pulses. Pulse velocity results of different systems should be compared with caution since measured values depend on the trigger principle [14].

ates in pitch-catch mode and is intended for thickness measurements and the detection of larger objects.

Fig. 12. Pulse-echo system for A-scan and B-scan measurements [16] Fig. 10. Transmission system for pulse velocity and attenuation measurements. Membrane shoes are developed for easier coupling [15]

In contrast, few commercial pulse-echo systems exist to date. One unit contained in a 19-box is depicted in Fig. 11. It enables digital A-scan and B-scan measurements as well as transmission experiments. Featuring a rectangular pulser with adjustable width, high gain/low noise amplifier, filtering, and time-gain control, this device is particularly suited for concrete testing.

Fig. 11. Pulse-echo system for A-scan and B-scan measurements

To achieve high axial resolution, usually largebandwidth transducers from 50 to 400 kHz center frequency are used for pulse-echo systems, most of them having bandwidths from 100 to 200 %. Acoustical coupling of transducers to the concretes surface is a major concern. The process of applying coupling agents such as honey or vaseline is time-consuming, and coupling fluctuations may exceed 6 dB. Rough surfaces need to be grinded, but still concave portions of the surface pose a problem to transducers with large diameter. Dry Point Contact (DPC) transducers avoid this problem by establishing a direct, point-like contact to the concrete without the need of a coupling agent [16]. Two versions with conical and flat tips have been developed. Fig. 12 shows a transducer array with 24 conical tip DPC-transducers, each 12 of them being interconnected as the sending and the receiving portion. The array, belonging to the pitch-catch echo system mentioned above, allows for fast coupling and scanning but seems to produce increased coupling fluctuations. Other detection approaches include a scanning Doppler vibrometer which enables scanning on a twodimensional grid [17]. Besides being an attractive means the signal-to-noise ratio needs to be improved. VII. APPLICATIONS To illustrate the range of ultrasonic concrete evaluation, a number of research topics and example applications are described in the following without

Meanwhile, a hand-held A-/B-scan unit is available from another vendor (Fig. 12). This system oper-

2002 IEEE ULTRASONICS SYMPOSIUM-744

Authorized licensed use limited to: IEEE Xplore. Downloaded on December 7, 2008 at 06:50 from IEEE Xplore. Restrictions apply.

striving for completeness. The list includes well-tried methods as well as results of current research. Compressive Strength The correlation of pulse velocity to compressive strength is among the earliest applications of ultrasound to concrete testing [1], [18]. This method relies on empirical correlation curves found in transmission experiments [Fig. 13(a)], which are subsequently used to estimate the compressive strength of concrete samples and elements. Unfortunately, there is no single relationship between these two quantities. Instead, a number of parameters have influence on the correlation. In Fig. 13(b), concrete age and curing conditions are added to the effects considered, thus making the estimation of compressive strength much more difficult. These problems were discovered soon [1], and the method should be applied only under conditions where most influential parameters are controlled.

Concrete Hardening Monitoring the hydration process of concrete can help to determine the time instant at which the concrete is able to take strain [21]. It is also used for recipe development. Often the pulse velocity measured in transmission is evaluated. Fig. 14 shows graphs of concrete with different admixtures. In another approach, the reflection coefficient of transversal waves is used, which is particularly sensitive since transversal wave cannot propagate in the fluid phase [22].

Fig. 14. Concrete hardening monitored with different admixtures [21]

Crack Detection Macroscopic cracks in concrete can be caused, e.g., by deterioration, overload, or earthquake. Depth measurement of surface-opening cracks is thus important for determination of the concrete condition. In order to do this, transducers are placed on either side of the crack, and the crack depth is estimated from pulse transit time. Because of the larger pulse velocity in steel than in concrete, reinforcement can cause interfering indications (Fig. 15). A number of methods have been proposed to handle this problem.

Fig. 13. Correlation of pulse velocity to compressive strength, (a) early-age concrete [19], (b) influence of concrete age and curing condition [20]

Fig. 15. Evaluation of crack depth: problems caused by reinforcement [23]

2002 IEEE ULTRASONICS SYMPOSIUM-745

Authorized licensed use limited to: IEEE Xplore. Downloaded on December 7, 2008 at 06:50 from IEEE Xplore. Restrictions apply.

Another task is the detection of cracks in posttensioned tendons. Calculations of guided ultrasonic propagation in a steel bar embedded in concrete have revealed that low attenuation can only be achieved for certain modes at narrow frequency bands (Fig. 16). Applying this result, it was possible to localize artificial tendon cracks in pulse-echo mode [24].

plitude detection can be used based on a threshold calculated from statistical noise distribution [27].

Fig. 17. SAFT-reconstruction of a pre-stressed concrete slab

Fig. 16. Attenuation dispersion curve of a steel bar in concrete [24]

SAFT-Imaging To overcome masking of object reflections by structural noise in single A-scans, SAFT (Synthetic Aperture Focusing Technique) reconstruction can be employed. The SAFT algorithm focuses A-scan measurements recorded along an aperture to each element of an image by coherent superposition, thus synthesizing a transducer of the size of the aperture with variable focusing [25], [26]. Linear apertures lead to two-dimensional images, planar apertures to three-dimensional images. Indications in SAFTimages are reconstructions of object boundaries and need to be interpreted using additional information. As an example, Fig. 17 shows the SAFT-reconstruction of a cross section through the bottom of a prestressed concrete bridge (see also Fig. 18, with indications explained). SAFT-reconstructions provide detailed information about the imaged concrete section and can therefore be used for detection and localization tasks. A present disadvantage is the tedious transducer coupling necessary at all aperture positions. Noise Modeling and Thresholding As has been pointed out before, structural noise is complicating the interpretation of ultrasonic A-scans and images of concrete. The recognition of indications within noise can be considered a detection problem. In an analogy to Radar detection methods, am-

For implementation, first a noise region is selected. Then the amplitude distribution of the noise is modeled by a parametric model using Weibull or LogNormal probability density functions. Given this distribution and a chosen false alarm probability, the threshold can be computed. Finally, all A-scan or image points with amplitudes below the threshold are discarded. Fig. 18 shows the SAFT-image of Fig. 17 after the thresholding operation. A Weibull distribution and a false alarm probability of 1 % were used. The result of investigations carried out so far is that noticeable noise reduction can be achieved without significantly changing the shape of indications. This method is also intended as a step toward quality control in image examination.

Fig. 18. SAFT-reconstruction of Fig. 17 after thresholding based on parametric noise modeling [28]

2002 IEEE ULTRASONICS SYMPOSIUM-746

Authorized licensed use limited to: IEEE Xplore. Downloaded on December 7, 2008 at 06:50 from IEEE Xplore. Restrictions apply.

Split Spectrum Processing Split Spectrum Processing is a signal processing techniques that exploits the different behavior of objects and concrete to frequency changes. The procedure consists of two steps. First, the measured signal is split in a filter bank. Then, all partial signals are exposed to a nonlinear operation such as polarity thresholding, and are superimposed to yield the processed output [30]. Fig. 19 shows examples of a measured and a processed signal. While being a promising approach, sensitivity against frequency-dependent sound attenuation remains a currently unresolved problem.

Being fundamental for the development of testing methods, a better understanding of ultrasonic propagation in concrete and scattering from objects is required. To achieve this goal, both measurements and modeling need to be performed. Additional progress is demanded in a number of fields. Most important are means for faster and more reliable transducer coupling. Robust measurement systems employing elaborated methods are needed, which offer unambiguous results to the user. Here a wide field for fundamental and applied research opens. IX. ACKNOWLEDGEMENTS The author is greatly indebted to all colleagues who have contributed information and figures, especially to S. Bussat, C. Groe, R. Jansohn, M. Krause, O. Kroggel, M. Lowe, K. Mayer, R. Marklein, J.S. Popovics, F. Schubert, S.P. Shah, V. Shevaldykin, T. Voigt, T. Yamaguchi, and D. Yuhas. Part of this work was supported by the Deutsche Forschungsgemeinschaft (German Research Council). Cooperation within the framework of the research initiative FOR 384 is particularly acknowledged (www.for384.uni-stuttgart.de). X. REFERENCES
[1] [2] E. Pohl, Prfung von Beton mit Ultraschall, Schriftenreihe der Bauforschung, Reihe Stahlbeton, Heft 4, Berlin: Deutsche Bauinformation, 1966, pp. 5-17. J. S. Popovics and J. L. Rose, A Survey of Developments in Ultrasonic NDE of Concrete, IEEE Trans. Ultrason. Ferroelec. Freq. Control, vol. 41, pp. 140-143, 1994. N. G. Zoldners and J. A. Soles, An Annotated Bibliography on Nondestructive Testing of Concrete, 1975-1983, in Proceedings, In Situ/Nondestructive Testing of Concrete, Detroit: American Concrete Institute, SP-82, 1984, pp. 745-825. M. Schickert, Ein empirisches Modell der frequenzabhngigen Schallschwchung von Beton," in Proceedings, DGZfP-Jahrestagung 1998, Berlin, DGZfP, 1999, pp. 233-240. P. Atkinson and J. P. Woodcock, Doppler Ultrasound and Its Use in Clinical Measurement, London, Academic Press, 1982. M. Schickert, Ein empirisches Modell der Dispersion von Ultraschall in Beton," in Proceedings, DGZfP-Jahrestagung 1999, Berlin, DGZfP, 1999, pp. 559-566.

Fig. 19. Split Spectrum Processing: examples of (a) a measured and (b) a processed signal [30]

[3]

VIII. CONCLUSION Due to its inhomogeneous composition, concrete is an interesting and demanding material for ultrasonic evaluation. Advanced methods are required for testing as well as understanding propagation characteristics. The described research results and applications show that ultrasonic testing can provide high potential solutions for the NDE of concrete. To promote regular use of ultrasonic NDE of concrete, more standards should be compiled and introduced. Up to date only standards on pulse velocity methods exist with the exception of [29].

[4]

[5] [6]

2002 IEEE ULTRASONICS SYMPOSIUM-747

Authorized licensed use limited to: IEEE Xplore. Downloaded on December 7, 2008 at 06:50 from IEEE Xplore. Restrictions apply.

[7]

[8]

[9]

[10] [11]

[12]

[13] [14]

[15]

[16]

[17]

[18] [19]

F. Schubert and R. Marklein, Numerical Computation of Ultrasonic Wave Propagation in Concrete Using the Elastodynamic Finite Integration Technique (EFIT), (these Proceedings). R. Marklein, K. J. Langenberg, R. Brmann, K. Mayer, and S. Klaholz, Nondestructive Testing with Ultrasound: Numerical Modeling and Imaging, Acoustical Imaging, vol. 22, 1996, pp. 757-764. F. Schubert and B. Khler, Numerical Modeling of Ultrasonic Attenuation and Dispersion in Concrete. The Effect of Aggregates and Porosity, in Proceedings, Nondestructive Testing in Civil Engineering (NDT-CE), Northampton: The British Institute of NDT, 1997, vol. 1, pp. 143-157. F. Schubert (Fraunhofer-Institute for Non-Destructive Testing, Branch Lab EADQ), private communication. J. Krieger, M. Krause, and H. Wiggenhauser, Tests and Assessments of NDT Methods for Concrete Bridges, in Proceedings, Structural Materials Technology III, Proceedings of SPIE, vol. 3400, 1998, pp. 258-269. M. Krause, H. Wiggenhauser, and J. Krieger, NDE of a Post Tensioned Concrete Bridge Girder Using Ultrasonic Pulse Echo and Impact Echo, Proceedings, Structural Materials Technology (SMT): NDE/ NDT Highways and Bridges, 2002 (to be published). S. Bussat, J. Neisecke (University of Dortmund, Germany), private communication. M. Schickert, Einflu der frequenzabhngigen Schallschwchung auf die Ultraschall-Laufzeitmessung an mineralischen Stoffen, in Proceedings, DGZfP-Jahrestagung 1994, Berlin: DGZfP, 1995, pp. 479-485. R. Long, M. Lowe, and P. Cawley, Investigation into Convenient Coupling for Ultrasonic Transducers when Inspecting Concrete Structures, Review of Progress in Quantitative NDE, vol. 19, New York: American Institute of Physics, 2000, pp. 1677-1684. A. A. Samokrutov, V. N. Kozlov, and V. G. Shevaldykin, Ultrasonic Defectoscopy of Concrete by Means of Pulse-Echo Technique, in Proceedings, 8th ECNDT 2002, Madrid: Spanish Society for NDT (AEND), CD-ROM, 2002, pp. 1-9. M. Krause and H. Wiggenhauser, Ultrasonic Pulse Echo Technique for Concrete Elements Using Synthetic Aperture, in Proceedings, Nondestructive Testing in Civil Engineering (NDT-CE) 1997, Northampton: The British Institute of NDT, 1997, vol. 1, pp. 135-142. RILEM NDT 1, Testing of Concrete by the Ultrasonic Pulse Method, RILEM, 1972. G. G. Carette and V. M. Malhotra, In Situ Tests: Variability and Strength Prediction of Concrete at Early Ages, in Proceedings, In Situ/Nondestructive

[20]

[21]

[22]

[23]

[24]

[25]

[26]

[27]

[28]

[29]

[30]

Testing of Concrete, Detroit: American Concrete Institute, SP-82, 1984, pp. 111-141. Y. Tanigawa, K. Baba, and H. Mori, Estimation of Concrete Strength by Combined Nondestructive Testing Method, in Proceedings, In Situ/Nondestructive Testing of Concrete, Detroit: American Concrete Institute, SP-82, 1984, pp. 57-76. C. U. Grosse and H.-W. Reinhardt, Ultrasound Technique for Quality Control of Cementitious Materials, in Proceedings, 15th World Conference on NDT, Rome, CD-ROM, 2000. T. Voigt, Y. Akkaya, and S. P. Shah, Determination of Early Age Concrete Strength by a Reflective Ultrasonic Technique, in Proceedings, 6th Int. Symp. High Strength/High Performance Concrete, Leipzig, 2002, pp. 1489-1501. T. Yamaguchi and T. Yamaguchi, Measurement of Thickness and Crack Depth in Concrete by Ultrasonic Methods, in Proceedings, NDT-CE Symposium 2000 and Seiken Symposium 26, Oxford: Elsevier, 2000, pp. 331-340. B. Pavlakovic, M. Lowe, and P. Cawley, Guided Ultrasonic Waves for the Inspection of PostTensioned Bridges, Review of Progress in Quantitative Nondestructive Evaluation, vol. 17, New York: Plenum Press, 1998, pp. 1557-1564. M. Schickert, M. Krause, and W. Mller, Ultrasonic Imaging of Concrete Elements Using SAFTReconstruction, J. of Materials in Civil Engineering (accepted for publication). M. Schickert, Towards SAFT-Imaging in Ultrasonic Inspection of Concrete, in Proceedings, Int. Symp. Non-Destructive Testing in Civil Engineering (NDT-CE), Berlin: DGZfP, 1995, pp. 411-418. R. Jansohn and M. Schickert, Objective Interpretation of Ultrasonic Concrete Images," in Proceedings, 7th European Conference on Non-Destructive Testing (ECNDT), Broendby, 7th ECNDT, 1998, vol. 1, pp. 808-815. M. Schickert, Hochauflsende Schnittbilder von Beton mit Ultraschall-Abbildungsverfahren, in Proceedings, Bauwerksdiagnose Praktische Anwendungen Zerstrungsfreier Prfungen, Berlin: DGZfP, CD-ROM, 2001, pp. 16. RI-ZFP-TU, Richtlinie fr die Anwendung der zerstrungsfreien Prfung von Tunnelinnenschalen, Bergisch-Gladbach, Germany: Bundesanstalt fr Straenwesen, 2001. N. Bilgtay, J. Popovics, S. Popovics, and M. Karaoguz, Recent Developments in Concrete Nondestructive Evaluation, in Proceedings, Int. Conf. Acoustics, Speech, and Signal Processing, 2001.

*Martin Schickert e-mail: [email protected]

2002 IEEE ULTRASONICS SYMPOSIUM-748

Authorized licensed use limited to: IEEE Xplore. Downloaded on December 7, 2008 at 06:50 from IEEE Xplore. Restrictions apply.