Cracks Classification using Acoustic Emission Technique on a Reinforced Concrete Beam

Summary

Claudio Guarnaccia of the University of Salerno presents a laboratory study using acoustic emission (AE) to detect and classify cracking in a reinforced-concrete beam during a four-point bending test. The broader motivation is to develop non-invasive methods for monitoring aging structures—especially bridges—by detecting damage early and potentially tracking it in service.

Approach and experimental setup

The general monitoring process described includes:

1. Reviewing the structure’s design and any previous test results, then visually inspecting it to identify likely critical areas.
2. Studying the materials and their current condition, since material properties affect how acoustic waves travel.
3. Selecting sensors and analysis settings based on the materials, background noise, and monitoring goals.
4. Installing and acoustically coupling the sensors to the surface. Sensor placement matters both for detecting events and for locating them through triangulation.
5. Validating individual sensors and communication among them before collecting data.
6. Applying an appropriate loading protocol, analyzing the signals, and using the results to locate and assess damage.

For the beam test:

• A reinforced-concrete beam was loaded monotonically in four-point bending.
• Three high-frequency resonant AE sensors were used. Their response peaked around 150 kHz, and they were spaced 50 cm apart following tests of wave propagation in the concrete.
• Strain gauges and displacement measurements were also used.
• The beam ultimately failed asymmetrically, with damage concentrated toward the right side.

AE signal measurements and crack classification

The presentation explains several measurements taken from an AE “hit”:

• Threshold: Separates detected signals from background noise.
• Amplitude: The signal’s maximum level.
• Duration: Time from the first to the last threshold crossing.
• Rise time: Time from the first threshold crossing to the maximum amplitude.
• Counts: Number of times the signal crosses the threshold.
• Energy: Integral of the signal’s envelope.

Two derived measures were used to classify cracking:

• Rise angle: Rise time divided by maximum amplitude.
• Average frequency: Counts divided by duration.

The study used a ratio of 1 to define the dividing line on a plot of average frequency versus rise angle:

• High average frequency and low rise angle were associated with tensile cracking.
• High rise angle and low average frequency were associated with shear cracking.

The presenter noted that published studies do not agree on a single best dividing ratio for concrete; values such as 1 and 10 have been reported. In this test, a ratio of 1 produced useful results.

Main findings

• AE activity differed among the three sensor channels, indicating that damage was not distributed symmetrically.
• By about 60% of the maximum load, localization analysis showed that most detected events were coming from the right portion of the beam. This indicated the eventual asymmetric failure location before the beam reached its maximum load.
• The left sensor (Channel 1) recorded relatively few shear-classified events, while the other channels recorded a mixture of tensile and shear events. This pattern was consistent with the beam’s asymmetric damage.
• As loading increased, the number of classified events rose. Near the end of the test, the signals indicated tensile cracking, shear cracking, and compression-related cracking near the top of the beam.

Conclusions and limitations

AE signal collection and analysis can help relate detected activity to structural damage, locate damage, and identify the type of loading or cracking occurring.

Dependable results rely on sensor placement, reliable sensor-to-surface coupling, material condition, threshold selection, and other analysis settings. Applying the method to operating structures is more challenging than using it in a controlled laboratory test, particularly in noisy environments such as bridges with traffic.

The research aims to support guidelines for applying AE monitoring to real structures, including prestressed-concrete bridges and steel structures. A robust, broadly accepted field-testing protocol remains a future goal.

Speakers and sources featured

• Speaker: Claudio Guarnaccia, associate professor of Applied Physics in the Civil Engineering Department at the University of Salerno.
• Equipment source: Vallen, which supplied the AE sensors and readout electronics.
• Research context: A consortium of structural-engineering laboratories in Italy funded the research.