Application Cases

Experiment Name: Experimental Study on Ultrasonic Attenuation in Concrete

Research Direction: Nondestructive Testing

Test Objective: To utilize piezoelectric smart aggregates for health monitoring of concrete materials, it is first necessary to study the propagation of acoustic waves in concrete based on embedded piezoelectric smart aggregates. Due to the relatively complex composition of concrete materials, including cement, fine aggregate, and coarse aggregate, stress wave propagation within them exhibits characteristics such as high attenuation, poor directionality, and complex propagation paths. Applying acoustic waves for health monitoring in concrete materials is much more complicated than in homogeneous materials like metal. Therefore, research on the propagation  of stress waves in concrete is necessary. Currently, most research on ultrasonic attenuation is based on externally attached sensors. However, the uncontrollable quality of the coupling interface can cause significant errors in ultrasonic attenuation measurements. Furthermore, the characteristics of the ultrasonic sound field are also related to the sensor shape. Consequently, it is necessary to study the attenuation of acoustic waves in concrete materials under the application of embedded piezoelectric smart aggregates.

Testing Equipment: ATA-2041 High-Voltage Amplifier, Waveform Generator, Charge Amplifier, Data Acquisition Card, Laptop Computer.

Experimental Procedure:
A concrete beam specimen was first fabricated, with dimensions of 2000 mm × 250 mm × 250 mm. Longitudinal reinforcement was also placed to withstand loads during movement and its own weight. An array of piezoelectric smart aggregates (SA) was arranged at the center of the beam cross-section, 125 mm from all sides, and numbered sequentially from 1 to 7. SA1 and SA2 were a pair of closely attached sensing devices; the two sensors were bonded together with adhesive and served primarily as a reference for attenuation signal analysis. The longitudinal spacing of transducers SA2 through SA7 was 400 mm, 400 mm, 400 mm, 200 mm, and 200 mm, respectively. The specific dimensions of the specimen and the sensor positions are shown in the figure below.

Figure: Specimen Specifications

The center frequency f of the ultrasonic pulse used in the experiment was set to nine values: 30 kHz, 50 kHz, 70 kHz, 90 kHz, 100 kHz, 110 kHz, 130 kHz, 170 kHz, and 190 kHz. Since the concrete beam contained six receiving  sensors and the experiment used narrowband pulses, the  excitation  frequency needed to be adjusted nine times throughout the entire experimental process. Each adjustment yielded six sets of data from sensors at different distances. For different frequency points, SA1 was excited, and each processing event acquired 8 columns of data information. Each data acquisition comprised 8 columns with 2×10⁶ rows. The first column was the time series, and the other seven columns were signal data, including the directly acquired original excitation signal from the first column and the data received by the other six channels after propagation through the concrete interior. This data volume could be rapidly acquired and processed using the data acquisition card, LabVIEW programs, and Matlab.

The experimental procedure is described in detail using the 100 kHz excitation frequency as an example. First, connect each SA and other instruments according to the diagram below. The power amplifier is connected to SA1, while SA2 through SA7 are connected to different channels of the charge amplifier. Correspondingly, the different output channels of the charge amplifier are respectively connected to different input channels of the data acquisition card, with each channel's data being independent.

Figure: Experimental System Diagram

Adjust the controls on the arbitrary waveform generator to read the Gaussian pulse. Since this arbitrary waveform generator does not have a built-in Gaussian pulse signal, it needs to be written using a Matlab function and imported into the arbitrary waveform generator in the .tfw file format. Read the written Gaussian pulse signal, adjust the waveform frequency to 100 kHz, and output it in the form of a single pulse. Adjust the output voltage to its maximum value of ±5 V.

Subsequently, adjust the power amplifier. Because the output voltage and power of the power amplifier are relatively high directly connecting it to the data acquisition card could easily cause damage. Therefore, it must be ensured that the output terminal of the power amplifier is only connected to the piezoelectric smart aggregate. First, adjust the resistance matching, then adjust the voltage gain factor. In this study, the gain factor was adjusted to 40 times, meaning the output voltage was ±200 V.

Check the port connections of the charge amplifier. The SA serves as the input to the charge amplifier, while the outputs are connected one-to-one to the data acquisition card via BNC cables. Circuit triggering allows for more precise control of the signal emission timing, which is beneficial for saving computational space. Therefore, a wire needs to be led from the PFI0 terminal of the data acquisition card and connected to the arbitrary waveform generator to facilitate trigger control through the LabVIEW program. The data acquisition card is connected to the computer via a USB cable, and signal transmission and reception control can be achieved through a self-written LabVIEW program.

Once the above connections are configured, signal acquisition can begin. The signal trigger is controlled through the LabVIEW program, upon which the arbitrary waveform generator immediately emits a 100 kHz Gaussian pulse. Set the sampling rate to the maximum of 2 M/s and the number of samples to 1200; each channel can then acquire a signal with a duration of 600 μs. The LabVIEW program automatically saves the data along with timestamps, and a single acquisition yields 6-channel data as shown in the figure below.

Figure: Received Signals from Sensors at Various Distances under 100 kHz Excitation Signal

Experimental Results:
One concrete beam specimen was fabricated, and seven piezoelectric sensors were embedded within it. To study the attenuation coefficient and attenuation rate at different distances, the sensor placement needed to ensure positional accuracy and directionality. The excitation scheme adopted was: signal excitation via SA1, reception via SA2 through SA7, with nine excitation frequencies selected between 30 kHz and 190 kHz, using Gaussian narrowband pulses. Data for 54 frequency-distance combinations were obtained from the 6 positions and 9 frequencies. From these data, the following conclusions can be drawn:

(1) Within the frequency range of 30 kHz to 190 kHz, the overall attenuation rate and the excitation frequency exhibit a quadratic polynomial relationship. there is a local minimum in attenuation at the frequency of 100 kHz. The study of the relationship between attenuation and frequency has guiding significance for selecting monitoring frequencies. Signal resolution reaches its optimum around 100 kHz, providing a reference basis for the selection of subsequent excitation frequencies.

(2) The attenuation coefficient increases with increasing distance, while the attenuation rate decreases with increasing distance. At a specific distance, the total attenuation rate varies with frequency; the total attenuation rate is only related to diffraction attenuation and is independent of material attenuation.

(3) Within the frequency range of 30 kHz to 190 kHz, the waveform correlation coefficient shows a trend of first increasing and then decreasing. A local maximum of the correlation coefficient exists around 100 kHz, indicating that using a 100 kHz signal for excitation yields the best waveform correlation.

Figure: ATA-2041 High-Voltage Amplifier Specifications and Parameters