Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
48 Each participant submitted a report within 2 weeks of test- ing, as per the instruction provided to them before testing. The results from both laboratory and field validation testing are presented and discussed in this chapter, as submitted by the participants. Field Validation Testing As shown in Figures 5.12 and 5.13 in Chapter 5, the Virginia bridge section was distressed at least by delaminations and vertical cracking. Results from validation tests on that bridge section by different NDT methods and devices are as follows. Impact Echo Five participants conducted IE tests on the Haymarket Bridge section. Results from these tests are compared in Figure 6.1. All the participants identified five sections of the deck with pre- dominant delaminations. Despite using various impact sources and different parameters (frequency, amplitude, thickness) for data analysis, the resulting contour maps and interpretations are generally in good agreement in defining the main areas of delamination. Comparing these maps with the observations made from the cores (Figure 5.13 in Chapter 5) confirms the capability of the impact echo method in detecting major delaminated areas. On each map, locations of the delaminated cores are marked with stars, and locations of the cores where no delamination was observed are marked with circles. Ultrasonic Surface Waves, Impulse Response, Infrared Thermography, and Chain Dragging and Hammer Sounding The ultrasonic surface waves (USW) data are presented in Fig- ure 6.2a. The primary objective of the USW test is to provide the condition assessment and quality of concrete through mea- suring concrete modulus. However, the presented modulus plot indicates that zones of very low moduli provide a good match with delaminated zones identified by other methods. Other technologies that primarily targeted delaminations were impulse response, infrared thermography, and chain drag- ging and drag/hammer sounding. Chain dragging and drag/ hammer sounding indicate delaminations at the correct loca- tions on the deck (Figure 6.2b through 6.2d). The impulse response technology was not very successful in detecting the delaminated areas. Infrared thermography was not as successful in identifying the delaminated areas. One of the reasons for this was that there were many people on the deck running various tests simultaneously. The shadows cast from these people could have affected the results. Therefore, it was difficult to draw a reasonable conclusion from the infrared thermography data. Ground-Penetrating Radar Five participants conducted GPR tests on this bridge section. Results from these tests are compared in Figure 6.3. All GPR maps are based on signal attenuation of the top rebar level. The delaminated areas are indirectly detected based on the areas with high-energy attenuation. Despite some discrepan- cies between maps regarding boundaries and the intensity of the deterioration, they are generally in good agreement. The five sections of the deck with predominant delaminations were also identified as deteriorated areas by all participants using GPR (Figure 6.3). These areas to some extent match those identified by the impact echo method (Figure 6.1). The depth of the concrete cover that was estimated by a participant using GPR is shown in Figure 6.4. Electrical Resistivity and Half-Cell Potential Electrical resistivity describes the corrosive environment, which in some cases is correlated to the corrosion rate, while the half-cell potential measurement yields the probability C h a p T e r 6 Results and Discussion (text continues on page 52)
49 Incipient Delamination Area with no DelaminationShallow Top Concrete Delamination A B C D E F G A B C D E F G 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 Figure 6.1. Comparison of results from IE tests on actual bridge section by different participants.
50 (a) (b) (c) (d) A B C D E F G 1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 Average Mobility 90-120 60-90 30-60 0-30 A B C D E F G 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 Stiffness 0.4-0.6 0.2-0.4 0-0.2 (e) Figure 6.2. USW (a), impulse response (b, c), infrared thermography (d), chain dragging and hammer sounding (e).
51 10 20 30 40 50 60 70 80 Longitudinal Distance (ft) Depth Corrected GPR Condition 0 5 10 -40 -30 -20 -18 -16 -14 -12 -10 -8 -6 -4 SERIOUS POOR GOODFAIR Signal Attenuation (Normalized dB) as Condition Indicator Figure 6.3. Comparison of results from GPR tests on actual bridge section by different participants.
52 10 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 10 5 0 20 30 40 Longitudinal Distance (ft) GPR-Estimated Concrete Cover (inches) 50 60 70 80 Figure 6.4. Concrete cover estimation based on GPR results. 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 Longitudinal Distance (ft) 5 10 5 15 25 35 45 55 65 75 85 95 Corrosion Rate Grade Very High H i g h Mod to Low Low Decreasing Corrosion Rate Indication 0 4 8 12 16 20 24 28 32 36 40 44 48 52 56 60 64 68 72 76 80 Longitudinal Distance (ft) 0 4 8 12 -720 -600 -480 -360 -307 -237 -180 -140 -100 -60 90% prob. of no corrosion 90% prob. of corrosion transition (a) (b) Figure 6.5. Electrical resistivity (a) and HCP (b). of active corrosion. Although these measurements represent different parameters, maps obtained from the two techniques correlate well (Figure 6.5). It must be noted that this is not always the case, and one technique cannot replace the other. Tables 6.1 and 6.2 show how well various participantsâ results were in agreement with the core conditions in terms of delamination detection. It should be again emphasized that this evaluation is not technology centered; rather, it is based on the participants. In these two tables, the red cells corre- spond to a falsely detected delamination and the green cells correspond to a correctly detected delamination. A cell is yel- low if the prediction was somewhat similar to the core condi- tion. NA means that the participant did not provide data for that particular core. All technologies report a number of false readings; however, most major defects are detected. (continued from page 48)
53 Repeatability Measurements The participants chose a number of ways to report their repeat- ability measurements. Some of the participants described and illustrated the repeatability of the technology in terms of the measured raw data, whereas others described the repeatability after some degree of analyzing the raw data. A third group reported the repeatability after the interpretation of the results. Figure 6.6 is a sample of the results extracted from the partici- pantsâ reports. The top figure illustrates the repeatability of a series of impact echo measurements described in terms of the frequency spectra (analyzed data). The next figure illustrates the repeatability of the impact echo measurements in terms of the delamination characterization (interpreted data). Finally, the two sets of repeatability results for half-cell potential and electrical resistivity are based on the raw field measurements. Laboratory Validation Testing: Fabricated Bridge Deck Impact Echo Six participants conducted either the ground-coupled IE or air-coupled IE tests on the fabricated deck to primarily detect the delaminated areas simulated in the deck. The condition maps reported by the participants are shown in Figures 6.7 and 6.8. For comparison, the horizontal distribution of defects as-built in the deck is also shown in each figure. For example, the main features in Figure 6.7c correspond to the frequency of the peak amplitude across the tested surface. Cold colors in this figure correspond to a lowest-frequency response, which is an indication of delamination. Another way of pre- senting the data is the cloud plot as shown in Figure 6.7d. Inspecting the two plots simultaneously may help improve the interpretation of the results. The participantâs protocol is to mark regions that exhibit both dense data cloud formation up to 4 kHz (Figure 6.7d) and a dominant low frequency (Fig- ure 6.7c), which indicate the likelihood of a near-surface delamination. Regions that exhibit only sparse cloud forma- tion up to 4 kHz, yet do exhibit a dominant low frequency, indicate the likely presence of other types of degradation. There is a good agreement between the maps provided by the participants, especially in identifying the shallow delami- nated areas. Both the air-coupled IE and ground-coupled IE methods show an acceptable capacity for detecting shallow delaminations. Significant experience seems to be needed in deep delamination detection and characterization. Ground-Penetrating Radar Five participants used GPR with only one system mounted on a vehicle. GPR condition maps are depicted in Figures 6.9 and 6.10. Some of the GPR condition maps identify the main areas of delamination, but in general the detection is not good. The likely reason for that can be explained in the fol- lowing way. The artificial delamination in the fabricated deck Participant Technology C1 C2 C3 C4 C5 C6 C7 C8 9 IE 9 Chain dragging and hammer sounding 6 Air-coupled IE 7 IE 1 IE 2 Infrared Correct Detection False Detection Approximate Detection No data available: NA Table 6.1. Detectability of Delamination by IE and Chain Dragging and Hammer Sounding Methods Participant C1 C2 C3 C4 C5 C6 C7 C8 1 8 NA NA NA 9 4/5 4 Correct Detection False Detection Approximate Detection No data available: NA Table 6.2. Detectability of Delamination by GPR (text continues on page 60)
54 (b) (c) 8070605040302010 8070605040302010 8070605040302010 -660 -120 -80 -40-160-200-270-350-420-540 Longitudinal Distance (ft) Run 2 Run 3 Run 1 Half-cell Potential Condition Map - Line D Repeatability - VA Bridge 90% prob. of no corrosion 90% prob. of corrosion transition 8070605040302010 8070605040302010 8070605040302010 Sound Fair Poor Serious Longitudinal Distance (ft) Run 2 Run 3 Run 1 Impact Echo Condition Map - Line D Repeatability - VA Bridge (a) Figure 6.6. Repeatability measurement results. Spectral amplitude, air-coupled IE (a); ratings based on spectral frequency, IE (b); voltage, HCP (c). (continued on next page)
55 Corrosion Rate Grade 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 Longitudinal Distance (ft) Electrical resistivity (kohm*cm) - Control 1 6 7.5 5 15 25 35 45 55 65 75 85 95 Very High H i g h Mod to Low Low Decreasing Corrosion Rate Indication Electrical resistivity (kohm*cm) - Control 2 6 7.5 6 7.5 Electrical resistivity (kohm*cm) - Control 3 (d) 10 20 30 40 50 60 70 80 Longitudinal Distance (ft) LINE D - REPEATS: Depth Corrected GPR Condition 0 2 4 -40 -30 -20 -18 -16 -14 -12 -10 -8 -6 -4 SERIOUS POOR GOODFAIR Signal Attenuation (Normalized dB) as Condition Indicator (e) (f) Line D, Repeat Tests 0 1000 2000 3000 4000 5000 6000 0 10 20 30 40 50 60 70 80 Longitudinal Distance, ft M od ul us , k si 1st Run 2nd Run 3rd Run Series4 There is no signal at the 3rd repeat Figure 6.6. (continued) Repeatability measurement results. Resistivity, ER (d); signal attenuation, GPR (e); and modulus, USW (f).
56 (a) (b) (c) (d) Shallow Delamination Shallow Very Thin Delamination Shallow Severe Delamination Deep Delamination Vertical Cracking Rebar Corrosion Test Lines and Points 1 3 5 7 9 11 13 15 17 Figure 6.7. Impact echo condition maps: plan of fabricated deck (a), Participant 4 (b), and Participant 6 (c, d).
57 (a) (b) (c) (d) Shallow Delamination Shallow Very Thin Delamination Shallow Severe Delamination Deep Delamination Vertical Cracking Rebar Corrosion Test Lines and Points 1 3 5 7 9 11 13 15 17 Frequency [kHz] 4 6 8 10 12 14 # of Column 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 181 # o f R o w A B C D E F G SC=Shallow delamination SC SC SC SC SC 4.2in. 3~6in. 4in. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Longitudinal Distance 1 2 3 4 5 6 7 L a t e r a l D i s t a n c e Longitudinal Axis T r a n s v e r s e A x i s Figure 6.8. Impact echo condition maps: plan of fabricated deck (a), Participant 7 (b), Participant 10 (c), and Participant 9 (d).
58 (a) (b) Shallow Delamination Shallow Very Thin Delamination Shallow Severe Delamination Deep Delamination Vertical Cracking Rebar Corrosion Test Lines and Points 1 3 5 7 9 11 13 15 17 (c) (d) Figure 6.9. GPR condition maps: plan of fabricated deck (a), Participant 8 (b), Participant 4 (c), and Participant 5 (d).
59 Shallow Delamination Shallow Very Thin Delamination Shallow Severe Delamination Deep Delamination Vertical Cracking Rebar Corrosion Test Lines and Points 1 3 5 7 9 11 13 15 17 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 G F E D C B A N = Normal A = Anomaly but the cause of the anomaly is unknown D = Delamination C = Corrosion M = Microcracking (a) (c) (b) (d) Figure 6.10. GPR condition maps: plan of fabricated deck (a), Participant 1 (b), and Participant 9 (c, d).
60 was created through the placement of synthetic inserts and not through the penetration of moisture and chlorides that would create a corrosive environment leading to rebar corrosion. A corrosive environment that would foster concrete dete- rioration (and, in advanced stages, initiate delamination at the rebar level) manifests itself as a very high attenuation of the GPR signal. Therefore, employing GPR in this fashion, where attenuation-based deterioration is the key to identifying poten- tially delaminated zones, means nothing in this fabricated specimen. One of the participants was able to recognize plastic inserts as anomalies in the GPR B-scans but did not identify the same thing in the attenuation plot (Figure 6.10c and 6.10d). This is because the inserts happened to be thick enough and produced enough dielectric contrast to be directly imaged. Results of this testing also confirm that corrosion-induced delaminations in real bridge decks are not easily reproducible in the laboratory using inserts. As shown previously, corrosion and corrosion-induced delamination were detected during the field validation using other NDT methods. These areas of delamina- tion and highly active corrosion primarily existed in the areas where there was a highly attenuated GPR signal measured at the top rebar level. This confirms the indirect detection of delamina- tion by GPR in real decks through EM wave attenuation caused by a conductive concrete, instead of through imaging. Infrared Thermography The participant conducting passive infrared thermography tests used a FLIR T400 camera. Two sets of images from these tests at two different times and ambient temperatures are shown in Figure 6.11. Figure 6.11b (40 min after sunrise) shows much clearer images than those shown in Figure 6.11c (at about noontime) and provides information on the sever- ity of the delamination by the significant difference in color. The difference in color indicates that the time and temperature are critically important to successfully detect the delamination with infrared methods. In addition, the fact that the delamina- tion in this deck is simulated with plastic foam materials has to be considered. In heat capacity and heat conductivity, such materials are significantly different from air. Chain Dragging and Hammer Sounding, Half-Cell Potential, and Electrical Resistivity The condition maps based on chain dragging and hammer sounding, HCP, and ER are shown in Figure 6.12. Chain dragging was able to detect shallow delaminations except for the thin or small ones, but was not able to detect the deep delaminations. Half-cell potential could recognize some corrosion activity on the right side (upper right corner in Figure 6.12) of the embedded reinforcement but not on the left side (lower right corner in Figure 6.12). The reason for this is that before HCP testing, the embedded electrical connection to the left section of steel was broken. This did not allow an electrical connec- tion to be established with that portion of the reinforcement. Because the right and left sections of steel were not electri- cally continuous, an individual connection was required for each section in order to perform HCP. Finally, the resistivity map does not actually yield any information regarding the deck condition because the corrosion of the steel was not the result of a natural process, such as chloride and moisture ingress into the deck or concrete carbonation. Instead, the previously corroded rebars were placed in the deck. Surface Wave Methods Four participants reported their results from the surface wave methods. One participant mapped the variation in modulus along the depth (Figure 6.13). As anticipated, the variation in modulus is reasonably uniform except for the delaminated areas that show as degraded concrete. Three participants used the method to estimate the depth of the simulated cracks. One participant used the ultrasonic wavesâ time of flight and another participant used the SASW method. One participant used two alternative methods: surface wave transmission (SWT) and time-of-flight diffraction (TOFD). The results are presented in Table 6.3. The SWT method seems to be the most reasonable one. Laboratory Validation Testing: retrieved Bridge Deck This deck, as shown in Figure 5.19 in Chapter 5, was seriously distressed. The deck almost universally contained stress- induced cracks of different orientations and was extensively delaminated. Information from coring further confirmed the seriously distressed situation of the deck (see Figure 5.20). The main goal of the deck validation tests was to evaluate the detectability of various NDT methods when a deck is severely distressed. The condition maps for the retrieved bridge section, as reported by different participants, are shown in Figures 6.14 through 6.18. Some of the participants chose not to report their results on this deck while others used the results from multiple tests they performed to interpret the results. Table 6.4 and Table 6.5 show the comparisons (in the same format as for the Virginia bridge) between results reported by participants and the coresâ conditions in terms of delamina- tion detection. None of the participants could report with certainty that the deck was uniformly damaged. This may be because most methods rely on anomalies in the signal to deci- pher the condition of the deck. (continued from page 53)
61 (a) Shallow Delamination Shallow Severe Delamination Vertical Cracking Deep Delamination Rebar Corrosion Test Lines and Points (b) Fusion Image Infrared Image (c) Fusion Image Infrared Image Figure 6.11. Infrared images: plan of fabricated deck (a) and Participant 2 (b, c). Infra- red image (c) is taken from the opposite side.
62 Figure 6.12. Plan of fabricated deck (a), chain dragging and hammer sounding (b), HCP (c), and ER (d). Participant 9. Shallow Delamination Shallow Very Thin Delamination Shallow Severe Delamination Deep Delamination Vertical Cracking Rebar Corrosion Test Lines and Points 1 3 5 7 9 11 13 15 17 0 2 4 6 8 10 12 14 16 18 Longitudinal Distance (ft) 0 2 4 6 8 -660 -540 -420 -350 -270 -200 -160 -120 -80 -40 90% Probability of Corrosion Activity 90% Probability of No Corrosion ActivityTransition 0 2 4 6 8 10 12 14 16 18 Longitudinal Distance (ft) 0 2 4 6 8 5 15 25 35 45 55 65 75 85 95 Very High Decreasing Corrosion Rate Indication H i g h Mod to Low Low (a) (b) (c) (d)
63 Shallow Delamination Shallow Very Thin Delamination Shallow Severe Delamination Deep Delamination Vertical Cracking Rebar Corrosion Test Lines and Points 1 3 5 7 9 11 13 15 17 (a) (b) M odulus (ksi) 1 G F E D C B A 5500 5000 4500 4000 3500 3000 2500 2000 2 3 4 5 6 7 8 9 Longitudinal Axis Tr an sv er se A xi s 10 11 12 13 14 15 16 17 18 Figure 6.13. Plan of fabricated deck (a) and apparent modulus mapping for delamination (b). Participant 10. Table 6.3. Comparison of Crack Depths (in.) Estimated by Participants and As Built Crack Ultrasonic SASW SWT TOFD As Built CK1 Down to top rebar 6.0 2.0 ~ 2.5 NA 2.5 CK2 Surface 4.8 1.1 ~ 1.5 NA 2.5 CK3 Between two rebar mats 4.8 2.5 ~ 3.7 3.8 3.0 CK4 Down to top rebar 5.4 4.0 ~ 5.7 5.8 6.0 CK5a Down to top rebar Shallow NA NA 2.5 Note: NA = not available. a Natural crack from the edge of the deck (depth was measured at the edge).
64 Figure 6.14. Impact echo condition maps: Participant 4 (a), Participant 6 (b), Participant 1 (c), and Participant 7 (d). (b)(a) (d)(c) 1 2 3 4 5 6 7 8 E D C B A D = Delamination C = Corrosion M = Microcracking N = Normal A = Anomaly but the cause of the anomaly is unknown 1 2 3 4 5 6 7 8 E D C B A A = Initial delamination D = Progressed delamination C = Corrosion M = Microcracking N = Normal # o f R o w Frequency [kHz] # of Column 4 6 8 10 12 14 A B C D E 2 3 4 5 6 7 81 SC SC SC SC
65 1 2 3 4 5 6 7 8 E D C B A N = Normal D = Delamination C = Corrosion A = Anomaly but the cause of the anomaly is unknown M = Microcracking 1 2 3 4 5 6 7 8 E D C A A A A A B A A A A A A A A A A A A A A A N = Normal A = Anomaly but the cause of the anomaly is unknown D = Delamination C = Corrosion M = Microcracking (a) (b) (c) (d) Figure 6.15. GPR condition maps: Participant 8 (a), Participant 4 (b), Participant 5 (c), and Participant 1 (d).
66 Fusion Image Infrared Image (a) (b) (c) (d) Figure 6.16. Condition maps of chain dragging and hammer sounding (a); HCP, Participant 9 (b); and infrared thermography, Participant 2 (c, d). 1 2 3 4 5 6 7 8 E D C B A N = Normal A = Anomaly but the cause of the anomaly is unknown D = Delamination C = Corrosion M = Microcracking Figure 6.17. Impact echo condition maps: Participant 9.
67 Figure 6.18. GPR condition maps: Participant 9. Participant Technology/Device Coring Location 1 IE/Ultrasonic 4 IE Scanning 6 Air-coupled IE 7 Air-coupled IE 9 Air-coupled IE 2 Infrared (fusion image) 9 Chain dragging and hammer sounding False Detection Approximate Detection B1 B2 B3 B6 C3 D2 D4 Table 6.4. Delamination Detectability by Impact Echo, Infrared, and Chain Dragging and Hammer Sounding Participant Technology/Device Coring Location B1 B2 B3 B6 C3 D2 D4 1 GPR 4 Aladdin 2 GHz 8 Geoscope 3 GHz 5 RIS Hi BrigHT 9 2.6 GHz False Detection Approximate Detection Table 6.5. Detectability of Delamination by GPR
