KSCE Journal of Civil Engineering (2012) 16(3):388-397 DOI 10.

1007/s12205-012-1295-0

Structural Engineering

www.springer.com/12205

Non-Destructive Testing Methods to Identify Voids in External Post-Tensioned Tendons

Seok Been Im* and Stefan Hurlebaus**
Received August 18, 2010/Revised June 6, 2011/Accepted June 23, 2011

Abstract
A considerable number of Post-Tensioned (PT) bridges have been constructed because PT systems enable them to carry significant traffic loads and have an aesthetical structure. However, strand corrosion has been a long-standing issue because it may lead to the failure of tendons and the deterioration of structural performance. The corrosion typically occurs in voided locations with exposed strands; thus, the inspection of voids in external PT tendons is important and necessary in order to protect strands before corrosion occurs. Based on literature review, several Non-Destructive Testing (NDT) methods are compared for effectiveness of identifying voids in external PT tendons, and the Impact Echo (IE), ultrasonic, and sounding inspection methods are then selected and assessed using small-scale and mock-up specimens. From the experimental results, the wave-based inspection methods, IE and ultrasonic methods, are difficult to apply in the field because the imperfect bonds between ducts and grouts obstruct the transmission of waves. However, the sounding inspection method is not affected by the discontinuities and successfully identifies voids in test specimens. Thus, the sounding inspection can be an effective tool for identifying voids because of its easy application in the field. Keywords: post-tensioned bridge, void, external tendon, impact echo, ultrasonic, sounding inspection

1. Introduction
Post-Tensioned (PT) structural elements are used quite often in bridges due to their ability to economically span long widths while providing an aesthetically pleasing structure. In addition, the use of PT systems in long span bridges has become influential as they reduce the overall cost of the substructure. This is especially significant when substructure costs increase due to complexities in span conditions such as bridges spanning seas, valleys, and urban areas. PT systems are also preferred in bridge construction because they greatly increase structural capacities and are fairly easy to implement effectively. The construction of a basic PT system consists of placing steel prestressing tendons in ducts which are entirely or partially embedded in concrete structures. Once the concrete element has hardened, the tendons are tensioned and as a result an overall compression force is exerted on the concrete. The compressive stresses induced by the PT tendons are advantageous since they reduce or perhaps even eliminate tensile stresses, which concrete has little strength to resist. These systems make it feasible and more economical to use concrete for long span bridge structures. PT systems can be used not only for new bridge construction but also for the rehabilitation and strengthening of existing bridges. PT bridge systems can include both external and internal post-

tensioned tendons. While the internal PT tendons are embedded in concrete structures with sheath ducts, the external PT tendons are installed inside box girders with High-Density Poly-Ethylene (HDPE) ducts. The external PT tendons are typically utilized to reduce the congestion between reinforcing steel and internal tendons. Along with reducing congestion external tendons are also advantageous since they are easier to inspect and replace than the internal PT tendons. The advantages of PT structures may not be realized if they do not achieve their designed service lives, which is typically over 75 years. Exposure to harsh environments contributes to the early deterioration of PT structures. In order to protect PT strands from early deterioration, they are placed in ducts and cementitious materials are grouted into the areas between the steel strands and the internal surface of the duct. The combination of the duct and the cementitious material provides a protective system for the strands against the most aggressive contributors to steel deterioration, corrosion. Although PT systems provide many advantages for designers and constructors, corrosion of the PT strands has raised many concerns about the longevity of these structural systems. Unlike conventional reinforced concrete systems, in which corrosion distress can be identified by the observation of excessive staining, cracking, or spalling of the concrete cover, corroding PT

*Member, Manager, Bridge and Structure, Civil Engineering Center, Samsung C&T Corporation, Seoul 137-858, Korea (Corresponding Author, E-mail: sbeeni75@gmail.com) **Associate Professor, Zachry Dept. of Civil Engineering, Texas A&M University, TX77843, USA (E-mail: shurlebaus@civil.tamu.edu) 388

Non-Destructive Testing Methods to Identify Voids in External Post-Tensioned Tendons

systems seldom show these surface distress indicators. These indicators are not visible in PT tendons since the tendons are embedded in ducts and are away from the external surface of the structure. The difficulty in identifying tendon distress is particularly problematic since the degree of corrosion is critical to the structural performance of PT elements. In 1967, the sudden collapse of the Bickton Meadows Footbridge in the UK raised concerns about the durability of PT bridges, while the failure of Yns-y-Gwas Bridge in Wales, UK, further emphasized the severity of corrosion in PT systems (Pearson-Kirk, 2003; Raiss, 1995; Schokker et al., 2001). Recently, the investigated failures of an external tendon in the Mid-Bay Bridge and a vertical loop tendon in the Sunshine Skyway Bridge once again called attention to deficiencies in the inspection and maintenance methods of grouted ducts in PT structures (Corven, 2001; Pielstick, 2002). The Virginia Department of Transportation (VDOT) reported failures in external PT tendons in the Varina-Enon Bridge in 2007. Consequently, they emphasized the need for an appropriate health monitoring system in PT elements based on a failure analysis (Sprinkel and Napier, 2008). The corrosion issue in PT systems has been identified in ducts with voids, and these voids in PT bridges are typically formed by entrapped air pockets, grout bleeding, and/or improper grouting (PCI, 1997; Im et al., 2010a; Trejo et al., 2009; Woodward and Miller, 1990). Recently conducted investigations in PT bridges reported that grouted external tendons commonly have voids inside the ducts (PearsonKirk, 2003; Woodward and Miller, 1990; Im et al., 2010b; ASBI, 2000; FDOT, 1999; FDOT, 2001; Hansen, 2007). Thus, current PT bridges may have voids in their PT element and it can be a potential source of severe corrosion under certain critical environmental conditions. While voids in grouted ducts can cause significant and long-standing issues for structural integrity of external PT tendon systems, little research has been conducted on how to effectively investigate these voids which may be a route for deteriorate sources that induce corrosion.

is generally performed to identify the existence of voids in ducts (Corven, 2001). Although the sounding inspection is subjective, it is easy to execute in the field and is a relatively rapid method for detecting voids in ducts. Also, this inspection is particularly effective in the field because it can be applied without a power supply. In addition, sounding inspection findings can be supported by the borescope inspection (Corven, 2001). 2.2 Impact-Echo Technique The Impact-Echo (IE) method uses mechanical impacts to generate stress pulses in concrete structures. This method has been studied by several researchers (Carino and Sansalone, 1990; Carino and Sansalone, 1992; Pessiki and Carino, 1988; Sansalone and Carion, 1989) and has yielded successful measurements for flaw detection. Because the impact generates a high energy pulse and can penetrate deep into the concrete structures, the IE method is particularly promising in the application of NDT methods. In addition, it produces a better signal-to-noise ratio than other ultrasonic techniques because of its low attenua uz et al., tion in composite materials such as concrete (Karaog 1998). Carino and Sansalone applied the IE method to detect simulated voids in grouted post-tensioning ducts which were located in a 3.3 ft (1 m) thick concrete wall specimen (Carino and Sansalone, 1992). Although the IE method can make measurements for flaw detection and thickness, it is difficult to produce an image of the structure as well as to directly apply in small grouted ducts such as external tendons. Moreover, the IE method is rather time-intensive for inspection of large specimens. To resolve this limitation, a movable IE scanner system was developed by Tinkey et al. (2005). The scanning system was applied to internal ducts. However, it was difficult to detect voids when the concrete cover was thicker than the diameter of the embedded ducts. Also, the IE scanner system is incapable of identifying voids when the ducts are partially filled with water (Tinkey et al., 2005). The authors then applied a movable scanning system to the mock-up slab and girder which included voided ducts simulated by pieces of Styrofoam (Tinkey and Olson, 2007). However, the scanning system could not detect voids well when the diameter of the ducts was small and the concrete covers were thick. 2.3 Ultrasonic Technique Ultrasonics is defined as the study dealing with sounds having a frequency of 18 kHz and higher (Biltz and Simpson, 1996). The ultrasonic technique can be used in concrete structures to identify concrete thickness (Krause et al., 1995), elastic modulus, and detect flaws such as cracks, voids, etc., (Schickert, 1995). Martin et al. (2001) performed tests on 1.6 inch (40 mm) internal grouted duct specimens with simulated voids by measuring the time-of-flight of transmitted pulses. These tests required the use of several receivers in a grid pattern on the surface of specimens. Although this technique depicted successful results for detecting voids in PT concrete beams, it is relatively time-consuming and

2. Non-Destructive Testing Methods

In recent years, a considerable amount of research on NonDestructive Testing (NDT) has been conducted to identify the locations and severity of damage in structures. Although much research have shown successful results in the laboratory, the application of NDT methods in the field is still challenging because of the noise generated during field measurements, confined and limited field conditions for measurement, and other reasons. Thus, an effective NDT method in the field needs to consider the following criteria: 1) reliable measurement, 2) high speed measurement, 3) ease of application, and 4) low cost. To provide a feasible NDT method for external PT tendons, a review of various NDT methods used in civil engineering applications is performed as follows. 2.1 Sounding Technique For the inspection of external PT ducts, a sounding inspection
Vol. 16, No. 3 / March 2012

389

Seok Been Im and Stefan Hurlebaus

costly because of the large number of transducers required. Iyer et al. (2003; 2005) evaluated the ultrasonic C-scan imaging system to verify the detection of corrosion and voids in specimens. The specimens were grouted in square PVC ducts with one strand in the center and scanning was conducted underwater. Two different levels of corrosion and voids in specimens were identified using the C-scan. During application, ultrasonic method contact transducers are typically used, and a couplant, a sticky resin, needs to be placed between the surface of specimens and the transducers. Thus, the use of the couplant or other substitutive materials is inconvenient when applying the ultrasonic method in the field. Although the application of ultrasonic techniques are limited in real structures, they can be promising if limitations can be resolved. Alternatives have been studied that involve noncontact ultrasonic systems such as air-coupled ultrasound (Martnez et al., 2003; Mendelsohn and Wiener-Avnear, 2002), point-contact sensors (Kozlov et al., 2006; Schickert et al., 2003; Schickert, 2005), or laser ultrasound (Kil et al., 1998; Scruby and Drain, 1990). Air-coupled ultrasound works typically in the frequency range of 50 to 100 kHz for application in concrete structures; however, it has the disadvantage of low signal amplitudes due to the impedance mismatch between the interfaces (transducer-air-specimen). Consequently, Blum (2003) studied increased signal amplitude using several air-coupled ultrasonic transducers combined into an array for generation of ultrasound. Recently, point-contact sensors were developed which do not require any coupling agent for ultrasonic measurements (Kozlov et al., 2006). To create better resolution and reduce the noise from the obtained signals, the ultrasonic method with point-contact sensors utilizes the Synthetic Aperture Focusing Technique (SAFT) (Schickert et al., 2003). Comparing with conventional transducers, however, it is still challenging to obtain reliable data (Schickert, 2005). Laser ultrasonics may be used for the generation and detection of elastic waves. These laser techniques provide numerous benefits over conventional piezoelectric techniques (Hurlebaus et al., 1998). Because optical techniques are non-contact techniques, they do not cause any errors associated with couplant thickness and/or applied pressure which is present when contact piezoelectric transducers are used. Couplants on the specimen surface can cause small changes in observed arrival times of surface acoustic waves. These small errors can become critical when the measured velocities are used in calculation for the localization of discontinuities. Another benefit of optical methods is that they are point techniques which may be used on any specimen regardless of size or geometry. A laser source provides the benefit of being broadband; that is, the resulting signal has components over a wide frequency bandwidth. Finally, using laser detection has the benefit of making absolute measurements with high fidelity (resonance free). However, this technique is sensitive to the surface condition of the test specimen. In addition, the use of laser ultrasonics is expensive and its application in the field is difficult due to laser safety requirements.

2.4 Magnetic Flux Leakage Technique The Magnetic Flux Leakage (MFL) technique is a promising technique in the NDT&E field because it is capable of detecting corrosion in embedded steels or metallic materials. This technique employs the use of magnetic field sensors called Superconducting Quantum Interference Devices (SQUIDS). These sensors are relatively sensitive and have a large dynamic range for inspection. Krause et al. (2001) studied the inspection of ruptured tendons in pre-stressed concrete structures using a SQUID system. It was reported that the MFL method was applied successfully to the detection of corrosion in external polymer ducts (Tilly, 2002). Although the MFL technique is effective in detecting corrosion in metallic materials, it cannot evaluate other discontinuities such as voids and cracks. To prevent tendons from corroding, it is necessary to identify the existence of voids before severe corrosion in tendons occurs. 2.5 Computerized Tomography Technique Using X-ray Computerized Tomography (CT) is a technique used to reconstruct images of an object projected by X-ray or Gamma ( )-ray. This technology is capable of obtaining any cross-sectional image of an object under projection, and the various intensities of the transmitting rays which are affected by the objects density reconstruct the cross-sectional image. It has revolutionized the medical diagnostic field and is also expanding quickly in industrial fields (Herman, 1980). The application of CT in concrete samples was performed by Martz et al. (1991). They applied the X-ray CT system to inspect voids and reinforcement bars in small concrete samples. In 1993, the authors studied the use of the -ray CT to perform more accurate inspections for voids and reinforcement bars in concrete specimens (Martz et al., 1993). In this research, small scale tendon samples including three strands and grouts were fabricated and scanned using the CT method. Fig. 1 shows a CT scanned image for the sample of grouted ducts with three strands. The image exhibits the existence of voids inside the ducts, but the size of voids can be affected by beam hardening near high density materials. Although CT produces detailed images of the tendon specimen, the device is expensive and requires accessibility to the concrete specimen from both sides which limits its application in the field. In addition, safety

Fig. 1. CT Scan Image of a Sample of Grouted Ducts with Strands

KSCE Journal of Civil Engineering

390

Non-Destructive Testing Methods to Identify Voids in External Post-Tensioned Tendons

Fig. 2. The Principle of Electrical Capacitance Tomography

Fig. 3. Small-Scale Test Specimen for Preliminary Test

requirements for the excessive exposure to radiation have to be resolved before applying the CT method in the field. 2.6 Electrical Capacitance Tomography Recently, a great amount of research using electric characteristics has been performed on NDT methods. Time-Domain Reflectometry (TDR) using electromagnetic waves has been studied for detecting voids around strands (Li et al., 2005). However, this technique requires access to strands in order to make the two-wire transmission line. Electrical Capacitance Tomography (ECT) has been studied and applied to reconstruct crosssectional images of oil flow inside the non-conductive pipe line (Ortiz-Alemn and Martin, 2005a). The ECT can obtain capacity data from multi-electrode sensors, and data by several thousands of iterations make the permittivity images in sections. These images can show the existence of air in oil flows (OrtizAlemn and Martin, 2005a). This application of the ECT is one of the most promising tomography techniques because it is safe, has a fast response rate, and is relatively inexpensive (OrtizAlemn and Martin, 2005b; Yang, 1997). Although this electrical method produces low resolution images of sections, optimized data processing enables the image to be enhanced (Gamio et al., 2005). Similar capacitance methods have been applied for detecting voids in HDPE ducts (Iaquinta, 2004). The investigators attached a pair of electrodes on a small-sized HDPE duct that included one strand, and obtained successful results for identifying voids and water-filled voids in ducts. Fig. 2 shows the principle of ECT sensors on HDPE ducts. However, the design of ECT is complicated when it comes to obtaining successful estimation; thus, careful design and thorough verification are required.

system consists of oscilloscope (TDS 3034B), power amplifier (Panametrics), and a pair of 50 kHz transducers (Panametrics). The IE method is performed using a Portable Impact Echo System (PIES), a commercial NDT package, and the testing system consists of iPaq computer, digitizer, battery, transducer, and impactors. The ultrasonic and IE methods are applied to the test locations from Lo1 to Lo11 (Fig. 3) and eight points at each location are examined along the perimeter of the duct. Figures 4 and 5 exhibit the output data in the frequency domain obtained from the IE and ultrasonic methods, but these NDT techniques are found to be ineffective in detecting voids in the small-scale test specimens. The figures show that there is no critical difference at the void location, Lo6. This is because the interface between the HDPE ducts and cementitious grout obstructed the elastic waves used in these techniques. In addition, the ultrasonics using contact transducers requires a couplant between the transducers and the surface of the specimens. This is a time-consuming process and probably impractical for use on long tendon systems in PT bridges. As a result, ultrasonic tech-

Fig. 4. Inspection Results Using Ultrasonic Method

3. Preliminary Testing on Small-Scale Specimens

Based on the literature review, the IE method, ultrasonic method, and sounding method are assessed using small and large scale test specimens. First, the IE and ultrasonic techniques are selected and assessed for preliminary testing. Secondly, the IE and sounding methods are tested using mock-up specimens representing a field condition. Sixteen small-scale test specimens are fabricated and tested using the IE and Ultrasonic methods (Fig. 3). The ultrasonic method is applied using the pitch-catch method, and the testing
Vol. 16, No. 3 / March 2012 391

Fig. 5. Inspection Results Using Impact Echo Method

Seok Been Im and Stefan Hurlebaus

niques are determined not to be advantageous to full-scale specimens.

4. Full-Scale Experimental Test Setup

External PT tendon systems are typically installed in a harped shape in the field because this can reduce or eliminate positive flexural moments acting on the mid-span and negative flexural moments at both end-spans in continuous span bridges. Based on the recent inspections of PT bridges by the researchers, voids usually exist in the external PT systems near end anchorage zones. To simulate void cases in the field, half of a harped tendon system is designed and fabricated with improper grouting. The schematic of the experimental specimen is shown in Fig. 6. First, the full-scale experimental setup is tensioned by 19 strands and 0.3% of their ultimate tensile strength (0.8 ksi, 5.52 MPa) is applied to straighten the strands and mimic field conditions. The strands are located inside a 4 inch (0.1 m) diameter duct made of transparent acrylic plastic. After stressing the strands, the interstitial spaces between the stands and duct were filled with Class A grout (cement and water with a watercement ratio (w/c) of 0.44) to consider the effect of bleeding. Artificial voids are then intentionally formed in the top anchorage zone and at five locations along the duct to simulate improper grouting procedures (Fig. 6). As shown in Fig. 6, the main void at the top anchorage zone and the five other small voids along the duct are connected by a top surface void line along the duct. These void lines are called bleed lines (also known as bug holes) and are formed by the evaporation of bleed water from the grout and entrapped air pockets (PCI, 1997). A total of 15 experimental specimens are tested for assessing the IE and sounding inspection methods.

Fig. 7. Testing Locations of Impact Echo Method

almost partially filled, and Lo2 is a small void which means that two or three top strands are exposed. Lo6 just has a bleed line indicating that strands are not exposed at all and may have a tiny gap due to shrinkage. Lo7 is the location of a local void which is intentionally formed to simulate a void due to the entrapped air. The selected locations are examined along the perimeter of the ducts. Figure 8 exhibits the results obtained from the IE system. The results appear to have irregular peak frequencies but it seems that there is no critical difference among different void sizes. This indicates that waves generated by impact are not passing through the duct because of tiny gaps between ducts and grouts, and the waves are transmitted around the ducts. The voided specimens are then repaired with high performance and non-shrinkage grouts, commercially available pre-packaged grouts. The grouts are injected from the grout port at the upper anchorage and they successfully fill voids along the ducts as passing Lo7. The IE test is then performed again at the same locations, Lo1, 2, 6, and 7. Figure 9 shows the results obtained from the IE system, and

5. Experimental Results: The IE Method

To identify the existence of tiny discontinuities on full-scale specimens, the IE method is applied to the specimens. The IE method utilizes the PIES system, and randomly selected locations along the entire specimen are examined. The IE method identifies the tiny discontinuities between ducts and grouts at the locations, and the following four locations in Fig. 7 are selected and tested for the in-depth inspection. From the figure, Lo1 and Lo2 are located in main voids, and Lo1 indicates a large void,

Fig. 8. Results Obtained from Full-Scale Specimen before Repair Grouting

Fig. 6. Full-Scale Experimental Setup Showing Void Locations

Fig. 9. Results Obtained from Full-Scale Specimen after Repair Grouting

392 KSCE Journal of Civil Engineering

Non-Destructive Testing Methods to Identify Voids in External Post-Tensioned Tendons

Fig. 10. Comparison of Results Affected by Repair Grouting

Fig. 11. Unrolled Drawing of PT Ducts in the Void Mapping

the results at inspection points seem to have similar peak frequencies. Fig. 10 compares the results between before and after repair grouting at the location, Lo1, and the IE test obtains a critically different signal after completely filling voided parts. Therefore, the IE test can be employed to examine a perfect filling condition after repair grouting; however, it is incapable of identifying the difference between voids and tiny gaps inside PT ducts. Thus, it is difficult to use for investigating voids where exposed strands are included that may be a potential source of the strand corrosion. In addition, the IE system is difficult to apply even in laboratory conditions because external tendons easily vibrate due to other impacts. Hence, the IE system cannot record data under unnoticeable vibrations while measuring in the laboratory and field; thus, the IE technique might be difficult to apply in the field without bridge closure.

6. Experimental Results: Sounding Method

The sounding inspection method is assessed in the full-scale test specimens and a steel impactor is employed to identify the existence of voids in ducts. The presence of voids is identified by detecting a drummy sound while tapping. In this research, soundings are recorded at every inch in the surveyed intervals (1 ft, 0.305 m) of specimens. Although the assessment of the sounding is subjective and depends on the inspectors judgment, they are classified as drummy and solid sounds. To assess voids obtained from the sounding and visual inspections, the void profiles are mapped on scaled sheets called void mapping sheets. Drummy soundings are recorded on the unrolled drawing of PT ducts as shown in Fig. 11 and Fig. 12.

Visual inspections are performed through the transparent ducts to assess the accuracy of the sounding inspections. The void maps developed from the visual and sounding inspections are assessed in a program that computes areas on a scaled map, and the void area at every one-foot (0.305 m) interval is determined. The calculated void areas are then used to compare the effectiveness of the sounding inspection method. To evaluate the effectiveness of sounding inspections, statistical analysis is performed to determine the differences between sounding and visual inspections. Fig. 13 shows the initial void profiles found by the sounding inspection and the visual inspection before repairing the specimen. In the analysis, void maps obtained from visual inspections are considered to represent the actual existence of voids, as it was possible to confirm their existence through the transparent acrylic ducts. The differences between the void quantities from the sounding inspections and the visual inspections are shown in Fig. 14. The differences between the void quantities from the sounding inspections and the visual inspections were compared using the null and alternate hypotheses (H0 and H1) as follows: H0 : Sounding Inspection Visual Inspection = 0 H1 : Sounding Inspection Visual Inspection 0 (1)

where, Sounding Inspection Visual Inspection is the mean of the population

Table 1. The t-Test of Difference between Sounding and Visual Inspections for All Voids Difference Sounding-Visual Mean -1.0743

7.7295

Sample size t-test 360 -2.637

p-value 0.009

Fig. 12. Marking Voids on the Void Mapping Sheet

Vol. 16, No. 3 / March 2012 393

Seok Been Im and Stefan Hurlebaus

Fig. 13. Void Map of Specimen Table 2. The t-Test of Difference between Sounding and Visual Inspections for Main Voids and 5 Local Voids Difference Sounding-Visual Mean -0.1056

9.4742

Sample size t-test 134 -0.129

p-value 0.898

Fig. 14. Scatter Plots between the Void Areas Obtained from Sounding and Visual Inspection

for differences between void areas detected from the sounding inspections and visual inspections. The results of the t-test are provided in Table 1. From the analysis, the null hypothesis is rejected at the 0.05 level of significance; therefore, the voids identified from the sounding inspection do not accurately represent the existence of voids in ducts. However, a correlation analysis between sounding and visual inspections reveals a highly positive linear correlation (correlation coefficient is 0.906) between the actual voids in the test specimens and sounding inspections. In addition, this assessment includes the inspection of tiny discontinuities due to shrinkage or bleeding of grouts in which strands are not exposed. Thus, the ttest of difference between sounding and visual inspections is performed for void profiles of main voids and five local voids, and the results are provided in Table 2. The results exhibit that the null hypothesis is accepted for the 95% confidence limit.

Therefore, interpretations of the correlation and t-test for relatively large voids reveal that the sounding inspection is accurate at identifying the location and existence of a void which can be a potential source of the strand corrosion. To verify the effectiveness of the sounding inspection in the field, external tendons in a Texas bridge were selected and then assessed. Fig. 15 shows the void profiles using the sounding inspection obtained from a tendon in the field. The sounding inspection method is capable of distinguishing between the voided sound and solid sound while tapping, and it shows a similar void trend with experimental results. Also, the void locations examined by sounding inspections are directly identified using a borescope and verified to include voids inside the tendon (Fig. 16). While the sounding inspection may lack the ability to detect extremely small voids and requires experienced personnel, it can be an effective tool for inspecting voids because of its ease of application and relative accuracy.

7. Conclusions
In this research, wave-based NDT methods, the ultrasonic and IE methods, and the sounding inspection method, which is generally performed in the field, are evaluated using small-scale specimens and mock-up PT specimens. The grouted specimens include tiny discontinuities between ducts and grouts; thus, transmitted waves in the ultrasonic and IE methods are blocked at the discontinuities. Although the IE method can assess whether repair grouts, non-shrinkage grouts, successfully fill voids in the test specimens, it cannot distinguish relatively large voids from the specimen with discontinuities. On the other hand, the sounding
KSCE Journal of Civil Engineering

394

Non-Destructive Testing Methods to Identify Voids in External Post-Tensioned Tendons

Fig. 15. A Void Profile Using Sounding Inspection in a Texas Bridge

identify voids in external PT ducts. However, it is expected that the ECT method can be a promising tool for inspection of external PT tendons in future research because it is free of an imperfect bond between the duct and grouts.

References
ASBI (2000). American segmental bridge institute grouting committee: Interim statement on grouting practices, American Segmental Bridge Institute (ASBI), Phoenix, AZ, USA. Biltz, J. and Simpson, G. (1996). Ultrasonic methods of non-destructive testing, Chapman and Hall, London. Blum, F. (2003). A focused, two dimensional, air-coupled ultrasonic array for non-contact generation, MS Thesis, University of Stuttgart. Carino, N. J. and Sansalone, M. (1990). Impact-echo: A new method for inspecting construction materials. NDT and NDE for Manufacturing and Construction, Hemisphere Publishing Corp., New York, pp. 209-223 Carino, N. J. and Sansalone, M. (1992). Detection of voids in grouted ducts using the impact-echo method. ACI Materials Journal, Vol. 89, No. 3, pp. 296-303. Corven, J. A. (2001). Mid-Bay bridge post-tensioning evaluation Final report, Florida Department of Transportation (FDOT), Corven Engineering, Inc, Tallahassee, FL. FDOT (1999). Corrosion evaluation of post-tensioned tendons on the Niles channel bridge, Florida Department of Transportation (FDOT), Tallahassee, FL, USA. FDOT (2001). Sunshine skyway bridge post-tensioned tendons investigation, Florida Department of Transportation (FDOT), Parsons Brinckerhoff Quade and Douglas, Inc., Tallahassee, FL, USA. Gamio, J. C., Ortiz-Alemn, C., and Martin, R. (2005). Electrical

Fig. 16. Inside View of Duct obtained from a Texas Bridge using Borescope

inspection method may not identify a tiny void and need experience for the assessment, but it is capable of identifying relatively large voids, which include exposed strands, in the research. In addition, the sounding inspection method is adequate for employing in the field because it can be applied without a power supply and is simple to execute. While many NDT methods for inspection are sensitive to ambient vibrations resulting from wind and traffic loads, the sounding inspection is not affected by these field conditions. Thus, the sounding inspection can be employed as an effective inspection tool in the field to assess void locations. Although the sounding inspection can be effective in investigating voids in external PT tendons, it still has issues regarding subjective assessment. In order to better evaluate the soundings, a new instrument using a microphone needs to be developed and assessed in future research. Based on the literature review, the ECT needs more research for its design and assessment to
Vol. 16, No. 3 / March 2012

395

Seok Been Im and Stefan Hurlebaus

capacitance tomography two-phase oil-gas pipe flow imaging by the linear back-projection algorithm. Geofsica Internacional, Vol. 44, No. 3, pp. 265-273. Hansen, B. (2007). Tendon failure raises questions about grout in posttensioned bridges. Forensic Engineering: Civil Engineering News, pp. 17-18. Herman, G. T. (1980). Image reconstruction from projections: The fundamentals of computerized tomography, Academic Press, NY. Hurlebaus, S., Jacobs, L. J., and Jarzynski, J. (1998). Laser techniques to characterize the effect of geometry on acoustic emission signals. Journal of Nondestructive Testing and Evaluation, Vol. 14, pp. 2137. Iaquinta, J. (2004). Contribution of capacitance probes for nondestructive inspection of external post-tensioned ducts. 16th World Conference on NDT, Session: Civil Structures. Im, S. B., Hurlebaus, S., and Trejo, D. (2010a). Effective repair grouting methods and materials for filling voids in external posttensioned tendons. Transportation Research Record: Journal of the Transportation Research Board, accepted 2010. Im, S. B., Hurlebaus, S., and Trejo, D. (2010b). Inspection of voids in external tendons of post-tensioned bridges. Transportation Research Record: Journal of the Transportation Research Board, accepted 2010. Iyer, S. R., Schokker, A. J., and Sinha, S. K. (2003). Ultrasonic C-scan imaging Preliminary evaluation for corrosion and void detection in posttensioned tendons. Transportation Research Record, No. 1827, pp. 44-52. Iyer, S. R., Sinha, S. K., and Schokker, A. J. (2005). Ultrasonic C-scan imaging of post-tensioned concrete bridge structures for detection of corrosion and voids. Computer-Aided Civil and Infrastructure Engineering, Vol. 20, No. 2, pp. 79-94. uz, M., Bilgtay, N., Akgl, T., and Popovics, S. (1998). Defect Karaog detection in concrete using split spectrum processing. IEEE Ultrasonics Symposium, Vol. 1, pp. 843-846. Kil, H.-G., Jarzynski, J., and Berthelot, Y. (1998). Wave decomposition of the vibration of a cylindrical shell with an automated scanning laser vibrometer. Journal of the Acoustic Society of America, Vol. 104, No. 6, pp. 3161-3168. Kozlov, V. N., Samokrutov, A. A., and Shevaldykin, V. G. (2006). Ultrasonic equipment for evaluation of concrete structures based on transducers with dry point contact. NDE Conference on Civil Engineering, pp. 496-498. Krause, M., Maierhofer, C., and Wiggenhauser, H. (1995). Thickness measurement of concrete elements using radar and ultrasonic impulse echo techniques. Proceeding International Conference on Structural Faults and Repair-95, London, UK, Vol. 2, pp. 17-24. Krause, H. J., Wolf, W., Glaas, W., Zimmermann, E., Faley, M. I., Sawade, G., Neudert, G., Gampe, U., and Krieger, J. (2001). SQUID system for magnetic inspection of prestressed tendons on concrete bridges. Insight Nondestructive Testing and Condition Monitoring, Vol. 43, No. 7, pp. 458-461. Li, J., Akl, L., Hunsperger, R., Liu, W., Chajes, M., and Kunz, E. (2005). Time-domain reflectometry to detect voids in posttensioning ducts. Journal of the Transportation Research Board, CD 11-S, pp. 243-252. Martin, J., Broughton, K. J., Giannopolous, A., Hardy, M. S. A., and Forde, M. C. (2001). Ultrasonic tomography of grouted duct posttensioned reinforced concrete bridge beams. NDT&E International, Vol. 34, pp. 107-113. Martnez, O., Akhnak, M., Ullate, L. G., and Espinosa, F. M. (2003). A

small 2D ultrasonic array for NDT application. NDT&E International, Vol. 36, No. 1, pp. 57-63. Martz, H. E., Roberson, G. P., Skeate, M. F., Schneberk, D. J., and Azevedo, S. G. (1991). Computerized tomography studies of concrete samples. Nuclear Instruments and Methods in Physics Research Section B, Vol. 58, No. 2, pp. 216-226. Martz, H. E., Schneberk, D. J., Roberson, G. P., and Monteiro, P. J. M. (1993). Computerized tomography analysis of reinforced concrete. ACI Materials Journal, Vol. 90, No. 3, pp. 259-264. Mendelsohn, Y. and Wiener-Avnear, E. (2002). Simulations of circular 2D phase-array ultrasonic imaging transducers. Ultrasonics, Vol. 39, No. 9, pp. 657-666. Ortiz-Alemn, C. and Martin, R. (2005a). Two-phase oil-gas pipe flow imaging by simulated annealing. Journal of Geophysics and Engineering, Vol. 2, pp. 32-37. Ortiz-Alemn, C. and Martin, R. (2005b). Inversion of electrical capacitance tomography data by simulated annealing: Application to real two-phase gas-oil flow imaging. Flow Measurement and Instrumentation, Vol. 16, pp. 157-162. PCI (1997). Chapter 3: Fabrication and construction. PCI Bridge Design Manual, Prestressed Concrete Institute, Chicago, IL. Pearson-Kirk, D. (2003). The performance of post-tensioned bridge. Role of Concrete Bridges in Sustainable Development Proceedings of the International Symposium, pp. 129-140. Pessiki, S. P. and Carino, N. J. (1988). Setting time and strength of concrete using the impact-echo method. ACI Materials Journal, Vol. 85, No. 5, pp. 389-399. Pielstick, B. (2002). Grouting of segmental posttensioned bridges in America. Transportation Research Record, No. 1813, pp. 235241. Raiss, M. (1995). Post-tensioned concrete bridges: The UK debate. Concrete International, Vol. 29, No. 2, pp. 23-26. Sansalone, M. and Carino, N. J. (1989). Laboratory and field studies of the impact-echo method for flaw detection in concrete. ACI Special Publication, Detroit, SP-112, pp. 1-20. Schickert, M. (1995). Towards SAFT-imaging in ultrasonic inspection of concrete. Proceeding International Symposium of Non-Destructive Testing in Civil Engineering (NDT-CE), Vol. 1, pp. 411-418. Schickert, M. (2005). Progress in ultrasonic imaging of concrete. Materials and Structures, Vol. 38, pp. 807-815. Schickert, M., Krause, M., and Muller W. (2003). Ultrasonic imaging of concrete elements using reconstruction by synthetic aperture focusing technique. Journal of Materials in Civil Engineering, Vol. 15, pp. 235-246. Schokker, A. J., Breen, J. E., and Kreger, M. E. (2001). Grouts for bonded post-tensioning in corrosive environments. ACI Materials Journal, Vol. 98, No. 4, pp. 296-305. Scruby, C. B. and Drain, L. E. (1990). Laser ultrasonics: Techniques and applications, Adam Hilger. Sprinkel, M. and Napier, C. S. (2008). VDOT experience with grouted tendons in Varina-Enon precast segmental post-tensioned bridge. Virginia Concrete Conference. Tilly, G. P. (2002). Performance and management of post-tensioned structures. Proceedings of the Institution of Civil Engineers, Vol. 152, No. 1, pp. 3-16. Tinkey, Y. and Olson, L. D. (2007). Sensitivity studies of grout defects in posttensioned bridge ducts using impact echo scanning method. Transportation Research Record: Journal of the Transportation Research Board, No. 2028, pp. 154-162 Tinkey, Y., Olson, L. D., and Wiggenhauser, H. (2005). Impact echo

396

KSCE Journal of Civil Engineering

Non-Destructive Testing Methods to Identify Voids in External Post-Tensioned Tendons

scanning for discontinuity detection and imaging in posttensioned concrete bridges and other structures. Materials Evaluation, Vol. 63, No. 1, pp. 64-69. Trejo, D., Hueste, M. B., Gardoni, P., Pillai, R., Reinschmidt, K., Im, S. B., Kataria, S., Hurlebaus, S., Gamble, M., and Ngo, T. (2009). Effect of voids in grouted post-tensioned concrete bridge construction: Volume 2, Texas Department of Transportation (TxDOT),

Austin, TX. Woodward, R. J. and Miller, E. (1990). Grouting post-tensioned concrete bridges: the prevention of voids. Highways and Transportation, Vol. 37, No. 6, pp. 9-17. Yang, W. Q. (1997). Modelling of capacitance tomography sensors. Science, Measurement, and Technology, IEE Proceedings, Vol. 144, No. 5, pp. 203-208.

Vol. 16, No. 3 / March 2012

397