Bundesanstalt für Materialforschung und -prüfung

International Symposium (NDT-CE 2003)

Non-Destructive Testing in Civil Engineering 2003
Start > Contributions >Posters > Ultrasonic: Print

Experimental and theoretical investigation to optimize an ultrasonic phased array for low frequencies

F. Mielentz, M. Krause, R. Boehm, H. Wüstenberg, BAM Berlin (Germany)
K. Mayer, R. Marklein, University of Kassel (Germany)


A new phased array transmitting equipment for the ultrasonic low-frequency range is presented. Using this technique, time-controlled excitation of the transducers arranged in the array can shift, deflect and/or focus the sound beam in concrete structures.

Measurements on concrete test blocks with different thickness and aggregate size have demonstrated the possibilities of the sound field control using the phased array technique. Focusing and deflecting possibilities of the sound beam are tested using a laser vibrometer in through transmission technique. These tests have proved that the phased array transmitter unit developed by BAM may also be used for further application in pulse-echo mode.

At present commercially available probes with the nominal frequency of 200kHz are used for this array. The 3D-Model-Calculation based on point source synthesis allows to prove the suitability of the probes and to optimize their arrangement on the concrete surface. The program calculates the delay times necessary for the desired directivity pattern.

The influence of the concrete aggregate size and of the air-pores on the sound field is investigated by EFIT-modeling (Elastodynamic Finite Integration Technique) [1]. The synthetic EFIT-data are compared with the experimental results.

Using modeling the parameters of the ultrasonic array can be varied and tested before measuring. In this way it is possible to optimize the system and especially the construction of transducers for NDT of concrete elements.


Application of ultrasonic methods in the construction industry for non-destructive testing of building components made of mineral materials differs greatly from ultrasonic testing of metals. Significant material inhomogeneities in concrete cause strong sound attenuation by scattering and absorption, thus ultrasound frequencies below 200kHz must generally be used. This corresponds to a wave-length (l) longer than 2cm. Common ultrasonic low-frequency probes usually have transducer diameters (D) smaller than 6cm to allow attachment of probes to irregular concrete surfaces (not as smooth as concrete formwork) without difficulty. This D/l ratio is very unfavourable for ultrasonic testing and results in an almost non-directional radiation for longitudinal waves. Furthermore, an additional intensive non-directional radiation of transverse and surface waves takes place. These circumstances considerably impede the evaluation of conventional impulse-echo testing and have necessitated a special optimization process for ultrasonic testing in the construction industry [2].

A scanning technique based on the synthetic aperture principle has been proved to be suitable for practical use. Based on this method a measurement apparatus equipped with an array of 10 probes has been developed by BAM and used successfully in numerous measurements for structural investigation of concrete building components [3].

If pulses with different time delays are sent to the probes of such a phased array while transmitting, the sound field within the material can be controlled (Figure 1).

Fig 1: Sound beam control using a phased array.

Non-destructive material testing using ultrasonic phased array sensors has been successfully used for metals over many years. Using this technique, the sound fields of a phased array sensor can be adjusted to the actual test arrangement and the same probe can be used for many different tests. Modern phased array equipment has a large number of transmitter and receiver units enabling continuous deflecting and focusing of the sound beam [4], [5], [6], [7], [8], [9].

Fig 2: EFIT-modeling (Elastodynamic Finite Integration Technique) of the wave propagation in concrete. Four different snapshots:
a. t = 7.428E-6 s,
b. t = 3.417E-5 s,
c. t = 5.794E-5 s,
d. t = 9.954E-5 s

Theoretical calculations demonstrate that this technique can be used also in concrete building components. The influence of the concrete aggregate size and of the air-pores on the sound field is investigated by EFIT-modeling. Figure 2 shows a 2D-simulation of the wave propagation in four snapshots. A concrete aggregate size of 16 mm and an air-pore content of 2 % was taken as a basic for this calculation. In this example, the time delay was adjusted in a way that the sound beam was focused on x = 0.3 m and y = 0.5 m.

In the following an example of a phased array transmitter will be presented which uses time-controlled excitation of the transducers in the array to enable deflecting and/or focusing of the sound beam within the concrete building component.

Concept of the phased array transmitter apparatus

The phased array sensor consists of the ultrasonic low-frequency probes arranged in an array connected via a multicore cable to the transmitter units. A transmitter stage is needed for each probe which can be triggered externally. Transmitter impulse time delays are generated by commercially available PC boards. The circuit boards are connected through control lines and interface boards to the transmitter stages which emit suitably time delayed transmitter impulses Figure 3 gives an overview of the setup.

Common phased array sensors have a rectangular or quadratic transducer membrane composed of narrow rectangular elements which can be individually driven. Low-frequency phased array sensors in the frequency range of around 100 kHz suitable for concrete applications are currently not available.

As stated in the introduction, BAM has produced measurement equipment with an array of ultrasonic low-frequency probes. The probes have been sequentially used as transmitter and receiver so far, thus deflecting or focusing has not been possible. Using a computer control program and special electronics for sound field control, the probes should be able to be used for these investigations. However, the available probes had to be checked to see whether or not their use in a time-controlled phased array was practical.

Fig 3: Scheme of the setup with the phased array transmitter unit
Fig 4: Directivity pattern for two probe arrangements.
Fig 5: Array arrangement of ultrasonic probes.

A computer program based on point source synthesis was used to prove the suitability of probes and optimize their arrangement. For this purpose the probe array is divided into transducer sub-elements and the sound excitation reference point is calculated as the complex sum of the elementary wave fractions [10], [11]. Figure 4 shows the calculated directivity pattern for two different probe arrangements with the same focal point on the reference line. In the right-hand figure, a clear grating lobe can be seen at s = 280cm, which may cause misinterpretation in the measurement evaluation. Because of better suppression of the grating lobe in the left-hand figure, the staggered probe arrangement is preferred for use in the phased array apparatus (Figure 5).

Measurement on a concrete test block

Directivity pattern of phased array sensors are normally checked using semicylindrical specimens. Such specimens with dimensions and concrete composition as used in the construction industry were not readily available, therefore the measurements were carried out on available concrete building components. The probe transmitter array was attached to one side of the concrete specimen by means of a template. Directivity patterns were measured on the other side of the specimen using through transmission technique along a measurement line using a laser vibrometer. Vaseline was used as a coupling medium to connect the array. Figure 6 shows the measurement setup with a laser vibrometer used as an ultrasonic sensor. The concrete surface has been covered with a reflecting coat of paint or foil around the measurement line. The laser beam automatically scans at 0.5mm spacing along the measurement line by means of a computer-controlled mirror scanner. The measured amplitudes are shown as a function of distance in the results.

Fig 6: Phased array transmitter apparatus with laser vibrometer to measure the directivity pattern using through transmission technique (see Figure 3).

Results on the concrete test block and outlook

For the measurements a test block with maximum aggregate size of 16mm and a thickness of 450mm was used. Three cases were investigated equivalent to the calculations: without focusing, using focusing and focusing with a deflection angle of 13°. Figure 7 shows a comparison of measured and calculated directivity pattern of the array. The measured signals are filtered and rectified and each signal amplitude is evaluated. In the left-hand figure, the directivity pattern of synthetic EFIT-data can be seen. The synthetic data was calculated with an aggregate size of 16mm an air-pore content of 2%.

Fig 7: Calculated and measured directivity pattern.

In the right-hand figure, the measured amplitudes are shown. The top figure shows the directivity pattern without focusing where all probes within the array were driven simultaneously.

The middle figure illustrates the directivity pattern for a focused sound field. The ultrasonic amplitude is higher than in the unfocused case and the sound beam is significantly narrower. The natural focusing of the array produces a smaller sound beam as it was simulated by the calculation. Therefore, the effect of focusing is not as significant in the measurement as it is in the calculation. However, the cramping sound beam can be seen clearly.

The bottom figure displays the combined effect of deflecting and focusing the sound beam. The comparison of the diagrams in the bottom row of Figure 7 shows that the amplitude decrease in the measurement is not as significant as it is in the calculation.

The measured results clearly show a change in the directivity pattern of the probe array controlled by a time delay in the transmitter impulse.

In Figure 8 the amplitude distribution in a measurement area of x = 400mm and y = 700 mm for the three cases: without focusing, using focusing and focusing with a deflection angle is shown. The used measurement setup is shown in Figure 6. In the third picture of Figure 8 an increase amplitude is recognizable in two areas (coordinates: x = 50mm, y = 450mm and x = 350mm, y = 500mm).

Fig 8:
Directivity pattern of the probe array measured by laser vibrometer and using through transmission technique (measurement area: x = 400 mm, y = 700 mm).

Fig 9:
Calculated directivity characteristics of the probe array (measurement area 400 mm x 700 mm, compare Figure 8 right).

The calculation of these areas (Figure 9) is based on the point source synthesis. The increasing of amplitude in the calculation is recognized in the same areas as it is in the measurement. These lateral grating lobes result from the 2D-probe arrangement. If this will lead to consequences for practical applications, it must be subject of further investigations.

These tests have proved that the phased array transmitter unit which was developed by BAM is capable of focusing and deflecting the sound beam of a probe array in a concrete building component with a maximum aggregate size as used in the construction industry. Taking into account the inhomogeneity of concrete, a qualitative comparison of the results shows a good correlation between measured results and model calculations for directivity patterns.

It is planned to carry out measurements where the phased array transmitter unit will be applied to concrete building components of other aggregate size distributions and thicknesses. Additionally a concept for the application of the phased array transmitter unit for the pulse echo method will be developed in further research activities.


The support of this study by the Deutsche Forschungsgemeinschaft (German Science Foundation) via grant number FOR 384 is gratefully acknowledged.


  1. R. Marklein: Numerische Verfahren zur Modellierung von akustischen, elektromagnetischen, elastischen und piezoelektrischen Wellenausbreitung im Zeitbereich basierend auf der Finiten Integrationstechnik. Dissertation im Fachgebiet Theoretische Elektrotechnik der Universität GH Kassel (1997).
  2. Wüstenberg, H. und B. Rotter: Möglichkeiten und Konzepte für Ultraschall-Prüfköpfe im Bauwesen in: Schickert, G. (Hrsg.); Internationales Symposium Zerstörungsfreie Prüfung im Bauwesen, 27. Feb. - 1. März 1991 in Berlin, Berlin: BAM, DGZfP (1991) Tagungsbericht Teil 2, S. 488-504
  3. Mielentz, F., Milman, B., Krause, M. und H. Wiggenhauser: Praxiseinsatz eines Ultraschallecho-Verfahrens zur Strukturuntersuchung von Betonbauteilen in: DGZfP (Hrsg.); Fachtagung Bauwerksdiagnose - Praktische Anwendungen Zerstörungsfreier Prüfungen, 21.-22. Januar 1999 in München, DGZfP-Berichtsband 66-CD (1999) Poster 3
  4. Mielentz, F., Krause, M. und H. Wüstenberg (BAM, VIII.4): Entwicklung einer Gruppenstrahler-Sendeeinheit für Ultraschalluntersuchungen von Betonbauteilen In: DGZfP-Jahrestagung 06.-08. Mai 2002 in Weimar, DGZfP-Berichtsband BB 80-CD, Vortrag 44, Berlin (2002)
  5. Mielentz, F., Krause, M., Wüstenberg, H. (BAM, VIII.4) and H. Wiggenhauser: Development of a phased array transmitting equipment for ultrasonic testing of concrete In: Proceedings of the 11th International Symposium on Nondestructive Characterization of Materials, 24.-28. Juni 2002, Berlin, DGZfP-Berichtsband, Vortrag B10-2 auf CD, Berlin (2002)
  6. Erhard, A., Wüstenberg, H., Haufe, U. and W. Möhrle: Berechnung und Bau von Gruppenstrahler-Prüfköpfen in: Proceedings of the 3rd European Conference on Non-destructive Testing, Florence (1984) p. 156-170
  7. Schenk, G., Möhrle, W., Montag, H-J., Erhard, A. und H. Wüstenberg: Ultraschall Real-Time-Scanner für die Schweißnahtprüfung in: DGZfP (Hrsg.); DACH-Jahrestagung der Deutschen Gesellschaft für Zerstörungsfreie Prüfung, 29.-31. Mai 2000 in Innsbruck, DGZfP-Berichtsband 73, Berlin (2000) S. 659-668
  8. Azar, L. and S.-C. Wooh: Experimental Characterization of Ultrasonic Phased Arrays for the Non-destructive Evaluation of Concrete Structures, Materials Evaluation, Vol. 57 (1999) 2, p. 134-140
  9. Tränkler, H.-R. und E. Obermeier: Sensortechnik 1. edition, Berlin, Heidelberg: Springer-Verlag (1998)
  10. Boehm, R., Erhard, A. und H. Wüstenberg: Dreidimensionale Modellrechnungen zum Einfluss von zylindrischen Bauteilkrümmungen auf das Schallfeld von Gruppenstrahlerprüfköpfen in: DGZfP-Jahrestagung 06.-08.05.2002 in Weimar, DGZfP-Berichtsband auf CD, Berlin (2002)
  11. Kühnicke, E.: Elastische Wellen in geschichteten Festkörpersystemen 1. edition, Bonn: TIMUG e.V. (2001)
STARTPublisher: DGfZPPrograming: NDT.net