| |
| FHWA > Engineering > Pavements > Concrete > FHWA-HRT-04-127 > Appendix C |
|
Products and Research
|
APPENDIX C FIELD INVESTIGATIONFour field sites were investigated for validation of the HIPERPAV II system:
It is believed that the number of field sites evaluated will provide the minimum level of information necessary to meet the objectives of this effort successfully. However, future data from field sites could be used for local customization. This section describes the steps performed for validation of both the JPCP and CRCP pavement sites investigated. C.1 JPCP Sites InvestigatedJPCP sites investigated include a section on U.S. Highway 50 in Illinois and a bypass section of a farm to market road near Ticuman, Mexico. The selection of both sites was heavily weighted on the fact that extensive early-age and performance information is available for both of these sections. C.1.1 JPCP Section on U.S. Highway 50, IllinoisA field investigation and monitoring of a JPCP test section on U.S. Highway 50 near Carlyle, IL, was performed from August 7-10, 2001. Following is some background information on this section. C.1.1.1 Background InformationThe test sections were constructed and instrumented from April to August in 1986 on U.S. Highway 50 just west of Carlyle, IL. Early-age monitoring was performed by researchers at the University of Illinois at Urbana-Champaign.(103) These data were used to develop mechanistic design procedures for the Illinois DOT. The following tables and figures provide a brief summary of the published data. Table 41 gives a brief description of the JPCP test sections.
Section AA borders a bridge abutment on the west and is located 4.8 km from the other sections, which run continuously between IA to NA from west to east. The JPCP sections in this study were constructed with 6.1-m dowelled transverse joints. A 102-mm cement aggregate mixture (CAM) was used as the subbase. The concrete mix design was kept constant throughout the test sections and is shown in table 42.
Strength gain data were reported by Zollinger and Barenberg.(103) The reported centerpoint loading flexural strength at 28 days showed an average of 5.8 MPa, and the third point loading flexural strength at 28 days showed an average of 4.9 MPa. The average 28-day modulus of elasticity for all sections is 28,680 MPa. Table 43 contains traffic data for each test section between 1987 and 1999 provided by Illinois DOT. This table shows the accumulation of ESALs for all sections. In this analysis, the traffic is divided into vehicular categories and factors to calculate ESALs.
Table 44 compares the design ESALs versus the consumed ESALs as of 1999. All sections but section AA had surpassed the number of design ESALs by 1999. Sections MA and NA had nearly three times the number of design ESALs by 1999.
Table 45 gives friction and ride quality data for the test sections at different times between 1990 and 1999, from Illinois DOT records.
Illinois DOT performed falling weight deflectometer (FWD) tests on the test sections in 1994 and 1998. Table 46 gives the results for D0, deflection under the load, and LTE.
N/A = Not applicable (no cracks observed) Before construction, the modulus of subgrade reaction was determined at several locations along the site and reported by Zollinger and Barenberg. An average static k-value of 130 psi was observed, although no test date was reported.(103) In addition, historical condition survey data for the seven test sections were obtained from Illinois DOT and are presented in tables 47 through 53. In summary, each section remains intact, with the only visible distresses being a very small amount of transverse cracking and construction joint deterioration in several sections and some patching in section IA. The sections are performing well, considering the level of ESALs and the age of the pavements.
* Data provided by Illinois DOT
* Data provided by Illinois DOT
* Data provided by Illinois DOT
* Data provided by Illinois DOT
* Data provided by Illinois DOT
* Data provided by Illinois DOT
* Data provided by Illinois DOT C.1.1.2 Evaluation ActivitiesThe monitoring included the following activities:
The data collected from this experiment were used with early-age information available for this site to validate the long-term JPCP performance models in HIPERPAV II. A condition survey with photo and video documentation for all seven test sections was performed during the afternoon of August 7, 2001. The location and severity of each distress was documented. As in the historical condition survey data, few distresses were evident in the sections. The major distresses included a few slabs with transverse cracking and some low severity transverse joint spalling. The pavement has performed well, considering all sections except section AA exceeded the number of design ESALs by 1999, according to the original design. Sections AA, IA, JA, KA, and LA were tested with FWD at midslab for backcalculation of the modulus of subgrade reaction (k) and at transverse and longitudinal shoulder joints for LTE. Each test consisted of a seating drop followed by drops at three increasing loads, typically near 9, 12, and 15 kips. The approach and leave sides of transverse joints were tested. Table 54 provides a summary of center slab deflection bowls that have been normalized to 9 kips and are compared to historical deflections in 1994 and 1998.
Table 55 contains load transfer data and statistics for transverse joints for approach slabs.
Table 56 includes load transfer data for transverse joints and cracks for leave slabs. Statistical data is also included in the table. The values for the leave slabs are greater than the approach slabs.
T-type thermocouples were installed at three different depths (13, 76, and 127 mm) at sections AA, IA, and KA to monitor the temperature profile of the pavement. An additional thermocouple was installed at section AA at 241 mm to monitor the temperature at the bottom of the pavement. Measurements were taken manually to provide corresponding pavement temperature data during FWD and DEMEC testing. Figure 76 gives the temperature profile for section AA.
Figure 76. Pavement temperature profiles for Illinois JPCP evaluation. The gradient moves from slightly negative and increases throughout the day. The results for sections IA and KA mirrored the trends of section AA. A DEMEC caliper was used to measure the joint and crack movements by measuring the distance between metal studs epoxied to the pavement surface. Approximately five consecutive transverse joints were tested in each section. In addition, four longitudinal joints and all transverse cracks between the consecutive transverse joints were tested. Transverse cracks were included because they reduce the total movement at transverse joints by allowing expansion and contraction. Figure 77 shows the joint movement in section AA at different transverse joints.
Figure 77. Joint movement for section AA in Illinois JPCP evaluation. The downward trend indicates joints closing under higher temperatures and opening under cooler temperatures. All other sections showed the same trend. After 36 °C, no further movement is observed, possibly because of joint closure after that temperature. Table 57 gives linear trendline data for all sections and includes transverse joints, transverse cracks, and longitudinal joints. High r² values show that the joint and crack movements follow a linear trend.
*T.J. = Transverse Joint, L.J. = Longitudinal Joint A Dipstick profiling device was used to establish longitudinal and diagonal surface profiles for pavement slabs and to compute IRI in each section. Profiles were measured during both the morning and afternoon to determine possible environmental effects of curling and warping on the pavement slabs. Figure 78 shows a typical longitudinal profile for section AA after correcting for slope. A curling pattern at every 6-m interval corresponding to the slab length for this section is evident from this figure.
Figure 78. Longitudinal profiles for section AA, 253-mm thickness. Six 152-mm and six 102-mm cores were drilled. The following list describes the size and locations of the cores:
The 152-mm cores were tested for splitting tensile strength. The 102-mm cores were tested for Young's Modulus of Elasticity (E) and CTE. Thickness measurements were also taken from the cores and are compared to the design thickness in table 58.
C.1.2 JPCP Section on Ticuman Bypass, MexicoIn the past, the project team has closely followed the design, construction, and interim evaluations of a JCP section in Mexico. This section was part of a demonstration project to promote the use of PCC for highways in Mexico. In 1993, a cement producer in cooperation with the Mexican Secretariat of Communications and Transports conducted this demonstration JCP construction project. The project entailed rehabilitating an existing asphalt pavement structure with a JCP overlay. Pavement evaluation studies were initiated during and immediately after construction. Two years later, a performance evaluation was performed that included deflection testing, evaluation of the riding quality, and condition surveys. On the second evaluation, it was observed that the traffic patterns for that area of the country had changed drastically, subjecting the study section to an excessive increase in traffic loading. Some minor distresses were observed during that evaluation that later progressed as the traffic loads increased. Given that the Ticuman bypass possessed many of the characteristics desired for validation of the JCP long-term pavement performance models in HIPERPAV II, the project team decided to evaluate this section further after approval from FHWA. This section, describes the Ticuman bypass JCP section and summarizes the design, construction, and post-construction evaluations to date. C.1.2.1 Ticuman Bypass Preliminary InformationThe Ticuman bypass section is located on the Yautepec-Jojutla farm-to-market road in the State of Morelos. The section serves as a bypass to the town of Ticuman. Figure 79 shows the general location and layout of the road. The Ticuman bypass is oriented to the north and south and services a resort area, an aggregate quarry, and local farming and ranching operations. The project length is approximately 8.5 km that, for the most part, run along the edge of a ridge, down into a flood plain, and across a river.
Figure 79. Ticuman bypass project location. The Ticuman bypass is a two-lane road with 3.35-m wide lanes. The pavement structure is composed of a nondoweled JCP overlaying an existing HMA pavement. The slab thickness averages 233 mm and has a joint spacing of 4.5 m. The existing structure is composed of an HMA layer with an average thickness ranging from 50 to 80 mm. From a site inspection before construction, it was observed that the HMA was severely cracked in some areas. Cores extracted before construction showed an average thickness of 250 mm for the base layer. The base layer showed a plasticity index of 16.7 classified as GC under the unified classification system (UCS) or A-2-6 under the AASHTO classification system. The natural soil material showed a plasticity index of 17.6 with classification CL (UCS) or A-6 (AASHTO). The mix design used for this project had the following proportions:
Given that the existing HMA pavement was in very poor conditions, the traffic for this section was minimal. However, after construction of the JCP, the traffic patterns changed drastically, and significant traffic was generated. Table 59 summarizes the traffic conditions throughout the history of the road.
Annual temperatures for Ticuman were reported to range from 5 °C to 38 °C. Historic weather records for the days during construction were obtained from the meteorological center in Mexico (Comision Federal de Electricidad). The JCP was constructed from September 23 to October 12, 1993, with slip form paving equipment. Construction proceeded as scheduled with minimal delays. Table 60 shows the concrete compressive strength and stiffness statistics for the specimens taken at the Ticuman bypass (laboratory specimens from samples taken in the field, and core specimens extracted directly from the slab) tested after 28 days. Figure 80 shows the mean concrete compressive and flexural strength gain curves for the Ticuman bypass. This information is based on mean values from test results.
Figure 80. Mean concrete compressive and flexural strength gain curves for the Ticuman bypass. Table 61 shows the statistics for the flexural strength and the splitting tensile test results at 28 days for the Ticuman bypass.
Table 62 shows the summary statistics of the slab thickness, as obtained from PCC cores.
The minimum thickness, as specified in the pavement design for the Ticuman bypass, was 200 mm. In the case of this particular project, since the existing asphalt structure was not leveled up, the variability in slab thickness seems reasonable for a JCP overlay, considering the high level of roughness of the existing asphalt structure. C.1.2.2 Previous EvaluationsDuring and after construction, an extensive evaluation of the concrete pavement was performed, including joint cracking surveys and Dynaflect® deflection data collection. A cracking survey was performed on October 12, 1993. The survey consisted of making an inventory by visual inspection of those joints that were cracked. From this evaluation, cracks were observed to occur on average at every third joint (every 13.5 m). However, the spacing between cracked joints was reduced, as demonstrated in subsequent evaluations. During late December 1993 and early January 1994, pavement deflections were measured with Dynaflect equipment. These deflection measurements were taken at the midspan of the slab and at the joints to determine the pavement structural capacity and LTE between joints, respectively. In the case of the Ticuman bypass, load transfer between slabs is provided solely by aggregate interlock, since no load transfer devices were considered or installed during the pavement construction. The deflection measurements on which this analysis is based were taken during the winter, and although the temperature during measurement was not low, it was expected that during the summer this would be higher, and thus, joint openings would be smaller. During July 1995, Dynaflect deflections were taken at the slab center and edge. In addition, a traffic survey and PSRs were obtained. A distress survey also was conducted in July 1995. That survey concluded that 3.3 percent of the slabs in the project were suffering from some type of cracking or spalling distress. Concurrent with the deflection data collection, 24-hour traffic information was collected over a 6-day period from a portable weigh-in-motion (WIM) station on the Ticuman bypass, which provided directional volumes and axle weights. From this traffic data, an average daily traffic of between 800 and 1200 vehicles was estimated. C.1.2.3 JPCP Ticuman Bypass Evaluation in 2001As part of this project, an extensive pavement evaluation was conducted for the Ticuman bypass during August 21 through 24, 2001. The pavement evaluation studies included testing deflection with Dynaflect equipment, determining PSI under morning and afternoon conditions, monitoring joint openings, conducting distress surveys, and extracting PCC cores for characterizing PCC properties. An initial data reduction of the above collected information is described in the following sections. To perform detailed condition surveys, five sections were selected along the project. The stations for those sections are:
Since limited traffic control was available for this project, the above sections were selected based on visibility conditions. A minimum representative sample of at least 40 slabs per section was evaluated (180 m), for a total of 1 km. Other distresses observed included localized separation of the longitudinal joint. This was reportedly due to the fact that tie bars were not installed on some pavement sections, as was determined soon after construction. Some tie bar retrofitting also was observed on the areas with no tie bars. Some spalling problems that appeared soon after construction were repaired with PCC patches. Although some crack sealing was performed in 1995, no more maintenance was observed thereafter. The majority of the cracks had not been sealed at the time of this survey. In some localized areas, lane-to-shoulder dropoff also was observed. In addition to the detailed condition survey, a video documentation of the entire road was performed at a relatively low speed to capture the distress condition throughout the project. The distress survey indicates that 63 percent of the distresses are located on the northbound direction, with a total of 22.4 percent of slabs presenting some distress type, and 37 percent on the southbound direction, with 13.3 percent distressed slabs. The distress percentages on both directions agree with the load conditions for this road, which indicate a greater percentage of heavy vehicles on the northbound direction. Figure 81 shows a compilation of the three distress surveys performed at different dates through the history of Ticuman bypass. This figure demonstrates that longitudinal cracking is the predominant distress type. In addition, a considerably large number of slabs have a mix of longitudinal, transverse, and/or corner cracking. These slabs were reported as shattered slabs. Researchers believe that the lack of preventive maintenance has contributed significantly to the distresses observed on this road.
Figure 81. Number of distressed slabs per kilometer for the Ticuman bypass. In addition to the above distresses, a faulting survey was performed on both lanes. Faulting was considered positive when the approach slab was higher than the departure slab. Figure 82 shows a distribution plot of the faulting recorded for all the sections investigated. Most of the faulting observed in this figure is between 0 and 2.0 mm and ranges between −2.5 mm and +4.0 mm. From this evaluation, section 1 showed slightly higher faulting readings than the rest of the sections. This section also presented a greater occurrence of sealant damage and spalling problems. DEMEC studs were epoxied at an average of five to six joints and cracks to measure joint movement due to changes in temperature as a consequence of the diurnal cycle. Joint movement monitoring was performed in the morning and afternoon conditions during the days of August 22, 23, and 24. Monitoring of PCC temperature was achieved by installing thermocouple sensors at different depths throughout the slab depth. Figure 83 shows a distribution and cumulative plot of the joint opening at the Ticuman bypass. The average joint opening observed at the joints was in the order of 0.31 mm/°C. This opening varied from 0.18 to 0.39 mm/°C.
Figure 82. Faulting distribution for the Ticuman bypass.
Figure 83. Measured joint opening during August 22 to 24. Deflection testing was performed on August 22 from 2 p.m. to 6 p.m. on both lanes and under two loading conditions: at midslab to measure structural capacity, and at the joints to measure LTE. From this analysis, an average LTE of 93.0 percent was measured for the southbound direction; 90.2 percent was measured for the northbound direction. A comparison of the LTE analysis performed with 1993 data shows that the LTE in 2001 was much higher than the LTE observed in 1993 (70 percent in average). The higher LTE values measured in 2001 initially were believed to be a difference in temperature conditions for each testing period. However, the range of temperatures observed for 1993 was from 22 °C to 39 °C, and those observed during 2001 ranged from 29 °C to 32 °C. Although a wider range of temperatures was observed during the spring of 1993, a comparison of LTE versus temperature for that period did not show any significant trend in LTE. During winter of 1993, a deflection analysis indicated an average effective joint spacing of 7.9 m. On the other hand, the joint movement analysis for summer of 2001 showed that all joints are working joints; thus the average joint spacing corresponds to the design joint spacing of 4.6 m. Since longer slab lengths result in larger joint openings and therefore reduced LTE, it is believed that the higher LTE values measured during summer 2001 are a result of the effective joint spacing observed. Structural capacity for this road was performed in several occasions after construction of the JCP on the Ticuman bypass. The first evaluation was performed at the end of 1993. During that evaluation, extensive deflection testing was performed at midslab for most of the slabs. A second evaluation was performed 2 years later, in 1995. Finally, during the visit in 2001, Dynaflect deflections were taken again at midslab for this purpose. Tables 63 and 64 compare the deflections taken in 1993, 1995, and 2001. Deflections correspond to the sensor under the load (D1) and the farthest sensor from the load (D5), which is believed to represent more closely the soil support conditions. Higher deflections were observed in 2001 than in the previous dates. No deflection data were obtained for the northbound direction in 1993.
From a backcalculation analysis, a significant reduction in stiffness was observed in the HMA subbase layer after 1995. It is believed that lack of maintenance, along with the increasing traffic loads in this highway, may have contributed to reduced structural capacity. To obtain a measure of the ride quality for the Ticuman bypass, Mays meter equipment was used for determination of PSI. A rather low riding quality was measured for this section, with no definitive difference between the southbound and northbound lanes. To detect any changes in the profile of the pavement slabs due to curling and warping, Mays meter measurements were performed in the morning (6:30 a.m.) and afternoon (1 p.m.) conditions. Figures 84 and 85 show the PSI measurements for the morning and afternoon conditions. For the southbound direction, the afternoon readings appear to be significantly higher than the morning readings. However, for the northbound direction, the opposite trend is observed on most of the readings. In these two figures, the PSI measurements taken on summer of 1995 are presented. The PSI measurements taken in 1995 on the southbound direction are slightly higher than the readings taken in the morning of summer 2001. However, they were very similar to those taken in the afternoon of 2001. The PSI results for the northbound direction did not show any definitive trend from morning to afternoon conditions. Furthermore, no difference in PSI readings was observed from summer of 1995 to summer 2001.
Figure 84. Comparison of PSI ratings on the southbound direction (summer 2001 versus summer 1995).
Figure 85. Comparison of PSI ratings on the northbound direction (summer 2001 versus summer 1995). Twelve PCC cores were obtained for PCC strength, modulus of elasticity, and CTE testing. The following findings are presented from the site inspection visit on August 2001:
C.2 CRCP Sites InvestigatedC.2.1 CRCP Section at I-30 and I-35 Interchange in Fort Worth, TXBecause of its extensive network of CRCP pavements, the State of Texas was selected for identification of one of the two CRCP sites used for validation of HIPERPAV II. During the summer of 2001, TxDOT was undertaking the improvement of the U.S. Interstate (I)-35/I-30 interchange in Fort Worth, TX. As part of the improvement, a private contractor was constructing CRCP pavement on the approach to a bridge structure on the main lanes and on several access ramps. It was deemed this site could provide an excellent opportunity for validation of the HIPERPAV II module. On July 16, 2001, a section of CRCP pavement was instrumented and monitored for 72 hours. This section was part of one of the access ramps on the I-35/I-30 interchange. Once at the site, all the preconstruction information was collected, including pavement design, mix design, and materials. The typical cross section at the construction site is shown in figure 86. The concrete was being placed in two stages. The right lane and monolithic curb were constructed first, and tie bars were installed to place the left lane and curb in a second stage.
(1 ft = 0.3048 m, 1 in = 25.4 mm) Figure 86. Typical cross section for the CRCP instrumented section. To characterize the properties for the concrete at the site, information was collected on the mix design, admixtures, and additives used, as well as the coarse and fine aggregate properties and geological type. The mix design for this project is shown in table 65.
C.2.1.1 Instrumentation ProceduresOn July 12, 2001, the project team met with TxDOT and contractor representatives to identify the section to instrument. Due to repositioning of the paver, the next section to be placed was scheduled for Monday, July 16, 2001. During the days before placing the section to be monitored, the previously placed sections were surveyed to identify the typical crack spacing. In addition, a friction test was performed at the site (as described later in this section). From the previous sections placed, the average crack spacing measured was approximately 4.9 m. This survey helped determine the location for the crack inducers on the instrumented section. The crack inducers were spaced at approximately 4.9 m. Figure 87 shows the position of the crack inducers on the instrumented area.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
