Imaging defects in concrete structures using accumulated SIBIE

https://doi.org/10.1016/j.conbuildmat.2014.05.018Get rights and content

Highlights

  • Smoothing procedure, consideration of incident angle were applied to each SIBIE images.

  • The full SIBIE images from multiple measurement points were used for imaging defects.

  • Horizontal and vertical positions of defects were measured regardless of the impact duration.

Abstract

Images created by Stack Imaging Spectral amplitudes Based on the Impact-Echo (SIBIE) method are basically one-dimensional because a single frequency domain spectrum for waves reflected from a single point due to impact is used to create them. The method has a limitation for representing two-dimensional defects in a cross section of a concrete structural element. This study focused on defect detection and visualization in a concrete structural element using multiple impacts and accumulated SIBIE. An impactor was used to induce energy at multiple points positioned at prescribed intervals along the surface of a structure element. A modified SIBIE method was applied to the collected data at multiple points and the SIBIE images were accumulated to generate one image (an accumulated SIBIE image). As a result, the defect positions are successfully represented in the cross section of the structural element using a two-dimensional coordinate system.

Introduction

The impact-echo method, one nondestructive testing method available for evaluating concrete component condition in situ, has been widely applied by numerous engineers and researchers since Carino et al. [1] first proposed its application in this area in 1986. It has been used to: estimate concrete strength and thickness; detect internal defects; detect voids in grout used in prestressed concrete tendon ducts; and estimate crack depths. The accuracy of data acquired using this method is, however, easily affected by such issues as surface condition, shape, and cross sectional dimensions of the concrete structural component being examined and by the characteristics of the impactor. In addition to these issues, depending on the structural component investigated (e.g., slab) and element dimensions (e.g., slab thickness), the preferred frequency to detected is not always the highest amplitude peak. These are significant problems inherent to the impact-echo method. As a result, engineers involved in impact-echo testing and data analysis need to have adequate skills and experience to reach correct conclusions.

Over the years, researchers have studied ways to improve the effectiveness of the impact-echo method for examining concrete components. Zhu and Popovics tried to improve impact-echo measurements and visualize concrete defects using microphones [2]. The approach solved an inherent coupling problem between sensor (accelerometer) and the concrete component, but could not eliminate problems inherent to the impact-echo method. Ohtsu and Watanabe proposed the SIBIE (Stack Imaging Spectral amplitudes Based on the Impact-Echo) method for imaging the cross section of the test object using frequency spectra of the waveforms measured using the impact-echo method [3]. Ata et al. improved the SIBIE method by measuring waveforms at two points on either side of the impact point [4]. These methods have been shown to provide acceptable information on void locations in grouted prestressing tendon ducts by imaging impact-echo results.

Images created by SIBIE are, however, basically one-dimensional although they present a cross section. The method uses a single frequency domain spectrum of the surface motion normal to the plane of impact that is captured at a point near the impact point (Fig. 1). As a result, the method has limitations when used to represent defects in the cross section using two-dimensional coordinates. These limitations are also understandable because of the fact that the SIBIE images are basically symmetric and radial (Fig. 2). Tokai et al. [5] tried to improve locating vertical cracks by superposing SIBIE images taken from several points that cross a surface crack in a zigzag manner. This approach is, however, an application of SIBIE used to locate vertical cracks in concrete. Alver and Wiggenhauser [6], [7] suggested a method that superimpose SIBIE images obtained from each impact point along a scanned-line. The method used frequency spectra data independently and without modification. Therefore images produced using this method are not very smooth and some irregular vertical images are produced by the data.

In this study, simulated defects in a plain concrete beam were measured using the impact-echo method with multiple impacts and results were visualized using a modified and accumulated SIBIE approach. Impacts were applied at multiple points at prescribed intervals along the top surface of the beam. Resulting SIBIE images were modified by smoothing the data and making corrections for the incident angle. Images at the impact points were accumulated and reimaged to provide two-dimensional impact-echo results.

Section snippets

Impact-echo method

The impact-echo method for concrete involves measurement of wave motion when the surface is subject to impact at a point. The elastic waves propagate through the concrete and are reflected between the surface of the specimen (boundary surface) and air interface, such as a crack and a void. Understanding the nature of the wave response allows for both estimation of the thickness of the concrete member as well as the depth of the internal discontinuity. Fig. 1 conceptually shows the impact-echo

Modifications of SIBIE

The images created by the SIBIE method are basically one-dimensional because: (1) a single amplitude spectrum is used (the peak indicates the predominant frequency for waves undergoing multiple reflections); and (2) waveform collected is detected at one point (waveforms may be collected at one or two points near the impact). To create adequate two-dimensional SIBIE images of defects, a large number of test points are necessary.

To modify and improve the SIBIE method so that two-dimensional

Measurement system

The measurement system used in this study consisted of impactors of varying diameters (steel balls on rods), an accelerometer, and a data acquisition system (dynamic strain meter) shown in Fig. 9. The unidirectional accelerometer had a measurement bandwidth of 1–25,000 Hz. The sampling time for the dynamic strain meter was set at 1 μs. Measurements were made on a concrete beam having three simulated defects constructed using wood plates embedded at different depths in the cross section. Fig. 10

Outline and model of analysis

To verify the effectiveness of the proposed, modified, SIBIE method, the impact-echo tests was simulated using the finite-element method. Finite-element analysis based results were compared against those from measurements. Similar to the tested specimens, the model had three internal voids at different depths (Fig. 13). The main difference between the model and tests was that the modeled, internal, defects were air voids. Material properties used in the model were: density of 2300 kg/m3; Young’s

Conclusions

This study assessed an approach for visualizing internal defects within a concrete specimen using the impact-echo method by applying a modified and accumulated SIBIE approach. The proposed method involved the application of impacts at multiple points at specified intervals along the surface of the tested element. The modified SIBIE approach incorporates smoothing of the results and correction for measured incident angles, with individual SIBIE images of each point being accumulated to make a

References (9)

There are more references available in the full text version of this article.

Cited by (0)

View full text