U.S. Department of Transportation
Federal Highway Administration
1200 New Jersey Avenue, SE
Washington, DC 20590
202-366-4000


Skip to content
Facebook iconYouTube iconTwitter iconFlickr iconLinkedInInstagram

Federal Highway Administration Research and Technology
Coordinating, Developing, and Delivering Highway Transportation Innovations

 
REPORT
This report is an archived publication and may contain dated technical, contact, and link information
Back to Publication List        
Publication Number:  FHWA-HRT-16-010    Date:  March 2017
Publication Number: FHWA-HRT-16-010
Date: March 2017

 

Using Falling Weight Deflectometer Data With Mechanistic-Empirical Design and Analysis, Volume II: Case Study Reports

CHAPTER 4.

CASE STUDY 4: RIGID PAVEMENT EVALUATION AND OVERLAY DESIGN FOR STABILIZED BASE LAYER

 

PROJECT OVERVIEW

Project 05-0200 was located on I-30 in Saline County, AR. Test section 05-0218—a JPCP placed on a stabilized base layer—was selected for the rigid pavement rehabilitation case study. This section is representative of the following selection factors:

  • Rigid pavement.
  • Wet-nonfreeze climate zone.
  • Rural principal arterial—interstate functional class.
  • “Poor” pavement condition.
  • Gravel subgrade classification.
 

The test section was originally constructed in September 1993. In the LTPP Program database, this was referred to as construction number 1. Since the initial construction, the section has been rehabilitated twice. The first rehabilitation was in February 1997 (referred to as construction number 2), in which lane-shoulder and longitudinal-joint sealing was performed. The second rehabilitation occurred in December 2002, in which transverse joint and crack sealing was performed on the section and was identified as construction number 3. Based on LTPP Program core data, the typical cross section of test section 05-0218 consisted nominally of 8inches (203mm) of PCC (surface layer), 7 inches (178 mm) of a treated base (lean concrete), 4 inches 102 mm) of a granular layer, a woven geotextile (interlayer), and a subgrade layer.

PAVEMENT CONDITION/PERFORMANCE

Distress survey results, in terms of joint faulting and slab cracking, were available in the LTPP Program database from 1996 to 2007. According to the available records, only four joints had faulting greater than 0.1 inches (2.54 mm), which did not appear until 2001.

In terms of slab cracking, the studied pavement section was in poor condition. Table 60 presents the number of cracks identified for test section 05-0218 from 1996 to 2007. To convert the number of cracks to the percentage of slabs cracked, the assumption that only one transverse crack occurred within each slab was made. The section is 500 ft (152.5 m) long, and slabs are 15ft (4.6m) long, so there are approximately 33 to 34 per lane, with a total of 66 to 68 slabs. With 39cracks identified in 2007, it was approximated that more than 50 percent of the slabs were cracked.

Table 60. Transverse cracking distress data for test section 05-0218.
Survey Date Number of Transverse Cracks
Low Severity Medium Severity High Severity Total
11/14/1996 0 0 0 0
6/29/2000 8 7 0 15
7/25/2001 17 11 2 30
8/28/2002 11 19 3 33
10/07/2003 9 22 1 32
9/15/2004 9 23 1 33
5/23/2007 4 34 1 39

 

TEST SECTION DATA

The MEPDG program requires a significant number of inputs, particularly for a level 1 analysis. The required design data for test section 05-0218 were obtained from the LTPP Program DataPave database. The data were generally not complete for any specific test section within project 05‑0200; therefore, sometimes the data available from other sections of the project were used to define the necessary design inputs.

Deflection-Testing Data.

Deflection data for test section 05-0218 were available from the LTPP Program database for several years of testing, including 2001, 2003, and 2004. The data used for the backcalculation were from the most recent date because that was a better representation of the current conditions in the field. Deflection testing was conducted following the LTPP Program protocols.

Equipment

Deflection testing was conducted with a Dynatest® FWD (SN 8002-132).

Sensor Configuration

The sensor configuration used in the conduct of the FWD testing is presented in table 61.

Table 61. Sensor configuration for the FWD testing.
Configuration Sensor Number
1 2 3 4 5 6 7 8 9
Offset (inches) 0 8 12 18 24 36 48 60 -12

 

Number of Drops and Load Levels

Three load level targets—9,000, 12,000, and 16,000 lb (4,077, 5,436, and 7,248 kg) with fourdrops at each load level were performed, and the resultant data were recorded. Seating drops were also performed at each test point, but deflection data for those drops were not recorded.

Test Locations/Lanes and Increments

The location of the FWD testing is presented in table 62. The 2004 testing occurred from 12:30to 3:37 p.m. at 10 different locations along the section. The data from J1 were used to backcalculate k and E values, data from J4 and J5 were used to determine LTE and differential deflections, and data from J2 were used to determine the presence of voids under the slab. Tests were conducted at mid-panel, adjacent to the outside lane/shoulder joint at the corner and at mid-panel, and in the outer wheelpath on the approach and leave sides of the transverse joints. The distance between tests conducted at mid-slab was between 30 and 70 ft (9.15 and 21.4 m), whereas the tests carried out on the pavement edges at corners had intervals between 9 and 23 ft (2.7 and 7 m).

Table 62. Location of the deflection test on the slab.
LTPP Code Location
J1 JCP middle lane at middle panel
J2 JCP pavement edge at corner
J3 JCP pavement edge at mid panel
J4 JCP load transfer outer wheelpath at joint approach
J5 JCP load transfer outer wheelpath at joint leave

Temperature Measurements

Temperature measurements were taken using holes drilled in the pavement at time intervals of either 15 or 30 min during the deflection testing. In 1996, 1999, and 2001, the holes were at depths of 1, 4, and 7 inches (25, 102, and 178 mm). In 2003 and 2004, the holes were at depths of 0, 1, 2, 4, and 8 inches (0, 25, 51, 102, and 203 mm). Figure 62 through figure 67 show temperatures in the slab at the time of each FWD test, along with the prevailing weather conditions. The slab temperatures were extracted from the LTTP database. It should be noted that the temperature readings were taken on two different days in 1999, namely April 15 and April 19. The temperature profiles for those 2 days were put onto the same graph because they were only a few days apart and the weather was similar on those 2 days. As it can be seen in the figures, the temperatures throughout the PCC slab changed daily and seasonally.

The temperature measurements collected in 1996 were taken during November. The slab temperature difference for this month is nearly zero. This means the shape of the slab was defined by the construction gradient present at the time the slab set. The next series of slab temperature measurements occurred in April 1999. Two records were available for this month, one in the afternoon and one in the morning. In the morning, the sun was out, and in the afternoon, it was cloudy. Figure 63 and figure 64 are good examples of how solar radiation affected the temperature distribution in the slab—the temperature difference was positive in both figures and was highest at about 11:00 a.m. A positive temperature difference will cause the slab to curl downwards. The next measurements were taken in July 2001. The temperature difference was much higher in the afternoon for the month of July when compared with the same measurements taken in April, and it reached a high of 24.3 ºF (13.5 ºC) at 1:30 p.m. causing the slab to curl downwards. The next two measurements were taken in October 2003 and September 2004. The average temperature difference was 14.9 and 19.7 ºF (8.3 and 10.9 ºC) in October and September, respectively. Overall, the highest temperature difference was observed in July, and the temperature difference was larger in the afternoon.

Click for description
1 inch = 25.4 mm.
°F = 1.8 × °C + 32.

Figure 62. Graph. Slab temperatures for the FWD test performed on 11/14/1996—cloudy.

 

Click for description
1 inch = 25.4 mm.
°F = 1.8 × °C + 32.

Figure 63. Graph. Slab temperatures for the FWD test performed on 4/19/1999—sunny.

 

Click for description
1 inch = 25.4 mm.
°F = 1.8 × °C + 32.

Figure 64. Graph. Slab temperatures for the FWD test performed on 4/15/1999—cloudy.

 

Click for description
1 inch = 25.4 mm.
°F = 1.8 × °C + 32.

Figure 65. Graph. Slab temperatures for the FWD test performed on 7/26/2001—sunny.

 

Click for description
1 inch = 25.4 mm.
°F = 1.8 × °C + 32.

Figure 66. Graph. Slab temperatures for the FWD test performed on 10/8/2003—sunny.

 

Click for description
1 inch = 25.4 mm.
°F = 1.8 × °C + 32.

Figure 67. Graph. Slab temperatures for the FWD test performed on 9/16/2004—sunny/partly cloudy.

 

Material Properties Data

This section summarizes the data obtained from the LTPP database for test section 05-0218 regarding its subgrade, treated base, subbase, PCC material properties, traffic, climate, and depth of water table or stiff layer.

Subgrade

Nineteen subgrade samples were retrieved from the project 05-0200 as part of the LTPP Program. No specific soil classification data were available for test section 05-0218. Subgrade soil samples from sections 05-0217and 05-0219 were tested and classified as AASHTO A-2-4, and this soil type was assumed for test section 05-0218.

Laboratory resilient modulus testing was performed on 10 samples from section 05-0224 in December 2002. The test results are illustrated in figure 68, where BS** is the code name for the sample. No other sections had resilient modulus test information. As summarized in figure 68, the resilient modulus of the subgrade ranged from approximately 9,000to 19,000 psi (62,053 to 131,000 kPa) for a bulk stress between 8 and 28 psi (55 and 193 kPa).

Click for description
1 psi = 6.89 kPa.
Figure 68. Graph. LTPP Program data of laboratory resilient modulus testing for the subgrade.

 

Additional subgrade properties, including the sieve analysis are summarized in table 63. Gradation of the subgrade samples is shown in figure 69. The average moisture content was 2.2percent. Other properties (such as Poisson’s ratio and coefficient of lateral pressure) were assumed to be the default values for the soil type. It should be noted that the Atterberg limits for the samples taken from test sections 05-0217 and 05-0219 were very high and did not match the soil classification (AASHTO A-2-4); therefore, the default values in the MEPDG were used for these inputs.

Table 63. Gradation of the subgrade.
Sieve Size Average Percent Passing
3 inch 100.0
2 inch 100.0
1.5 inch 99.6
1 inch 98.5
3/4 inch 96.8
1/2 inch 91.9
3/8 inch 87.9
No. 4 77.5
No. 10 68.1
No. 40 57.3
No. 80 50.5
No. 200 42.1
1 inch = 25.4 mm.

 

Click for description
1 inch = 25.4 mm.
Figure 69. Graph. Gradation of the subgrade samples.

 

Granular Subbase

The granular subbase was identified as a crushed stone. The sieve analysis for the subbase layer is summarized in table 64; figure 70 illustrates the gradations of the subbase samples. The average moisture content was 3.5 percent. No information regarding Atterberg limits was available; therefore, the MEPDG default values based on material type were used. Laboratory resilient modulus testing results were available only for section 05-0213. The test was performed in December 1999 for 15 samples, with the test results illustrated in figure 71.

 

Table 64. Gradation of the granular base samples.
Sieve Size Average Percent Passing
3 inch 100.0
2 inch 100.0
1.5 inch 99.2
1 inch 91.6
3/4 inch 84.0
1/2 inch 72.6
3/8 inch 65.4
No. 4 49.4
No. 10 33.4
No. 40 16.0
No. 80 8.8
No. 200 5.1
1 inch = 25.4 mm.

 

Click for description
1 inch = 25.4 mm.
Figure 70. Graph. Gradation of the granular base samples.

 

Click for description
1 inch = 25.4 mm.
1 psi = 6.89 kPa.

Figure 71. Graph. Summary of the base aggregate resilient modulus from LTPP Program laboratory testing.

 

Based on the test results summarized in figure 71, the resilient modulus for this layer varied between 10,000 to 60,000 psi (68,900 to 413,400 kPa) for the bulk stress of 20to 100 psi (138 to 689 kPa). Considering the fact that resilient modulus testing was not performed on any samples from test section 05-218 and, moreover, recognizing the difficulty in determining the resilient modulus due to the unknown confinement condition, the typical resilient modulus suggested by the MEPDG for the granular layer of 30,000 psi (206,820 kPa) was used.

Other properties of this layer (such as Poisson’s ratio and coefficient of lateral pressure) were defined as the MEPDG default values for the material type.

Treated Base Layer

The treated base layer was a 7-inch (178-mm) lean concrete mixture. Sieve analysis test results were available for the coarse aggregate used in the lean concrete base. Six coarse aggregate samples were obtained as part of the LTPP Program data collection and were classified as quartzite. The results are presented in table 65 and figure 72. In addition to the coarse aggregate properties, the LTPP Program database contained the compressive strength, static elastic modulus, flexural strength, and the split tensile strength for the lean concrete base. The testing dates of these samples varied for each test. The compressive strength specimens were tested 4and 13 years after the section was constructed; however, the database did not specify whether the specimens were cast at the time of construction or sometime thereafter. Similarly, the static elastic modulus was tested 13 years after the initial construction date, but the age of the specimen was unknown. The flexural strength specimens were tested 2 years after the construction date. The split tensile strength was the only test that had a known age, which was 28 days. A summary of these tests for the lean concrete base is presented in table 66. The density of the cores was also measured. There was no information regarding the CTE of the lean concrete.

Table 65. Sieve analysis test results for the lean concrete base layer.

Sieve Size Average Percent Passing
3 inch 100.0
2 inch 100.0
1.5 inch 96.8
1 inch 68.2
3/4 inch 34.7
1/2 inch 5.3
3/8 inch 3.7
No. 4 2.8
No. 10 2.7
No. 40 2.2
1 inch = 25.4 mm.

 

Click for description
1 inch = 25.4 mm.
Figure 72. Graph. Comparison of gradation of the lean concrete base samples.

 

Table 66. Summary of lean concrete base laboratory testing results.

Laboratory Test Average Test Result
4-year compressive strength (psi) 7,331
13-year compressive strength (psi) 5,480
Elastic modulus (psi) 3,317,000
Modulus of rupture (psi) 522
Splitting tensile strength (psi) 629
Poisson’s ratio 0.29
Density (lb/ft3) 140.3
1 psi = 6.89 kPa.
1 lb/ft3 = 0.0160 g/cm3.

 

There was a significant amount of variability in the elastic modulus testing results (approximately 60 percent COV). Because the strength properties of the cement stabilized layer were defined through the elastic modulus in the MEPDG software and the very high variability from the test results, the typical value for a cement stabilized layer (2 million psi (13,780,000 kPa)) was selected for this parameter. There was also significant variability in the Poisson’s ratio results, so the default value (0.20) was also used for this property.

PCC Layer

The PCC layer had an average thickness of 7.8 inches (198.1 mm) based on available coring data. The LTPP Program database contained laboratory compressive strength, elastic modulus, Poisson’s ratio, modulus of rupture, and split tensile test results for PCC samples from project 05-0200. The testing dates of these samples varied for each test. The compressive strength specimens were tested 4 and 13 years after the section was constructed; however, the database did not specify whether the specimens were cast at the time of construction or sometime after. Similarly, the static elastic modulus and Poisson’s ratio were tested 9 and 13 years after the initial construction date, but the age of the specimen was unknown. The flexural strength specimens were tested 2 years after the construction date. The split tensile strength was the only test that had a known age, which was 28 days. These data are provided in table 67.

The average Poisson’s ratio for all the 05-200 sections ranged from 0.20 (9 years) to 0.24 (13 years), as summarized in table 67. However, the Poisson’s ratio measured for test section 05‑0218 in 2006 was reported as 0.15, which is the value typically assumed for PCC. Therefore, the value determined specifically for the section was used in the analysis. Similarly, the unit weight determined specifically for test section 05-0218 (138 lb/ft3 (2.2 g/cm3)) was used in the analysis, which was only slightly lower than the overall average.

Table 67. Summary of PCC laboratory testing results.
Laboratory Test Average Test Result
4-year compressive strength (psi) 8,189
13-year compressive strength (psi) 7,798
9-year elastic modulus (psi) 5,530,000
13-year elastic modulus (psi) 4,020,000
2-year splitting tensile strength (psi) 475
13-year splitting tensile strength (psi) 735
9-year Poisson’s ratio 0.20
13-year Poisson’s ratio 0.24
Modulus of rupture (psi) 623
Density (lb/ft3) 141.6
1 psi = 6.89 kPa.
1 lb/ft3 = 0.0160 g/cm3.

 

The laboratory-measured elastic modulus specific to test section 05-0218 was 3.6 million psi (24,804,000 kPa) from the 2006 data, which was slightly lower than the overall average. Using the correlation from the MEPDG between the elastic modulus and modulus rupture, for an elastic modulus of 3.6 million psi (24,804,000 kPa), a modulus of rupture of 630 psi (4,341 kPa) was obtained. This modulus of rupture estimated for the section was close to the overall project average.

No data were available on the thermal properties of the PCC. For the purposes of this analysis, it was assumed that the aggregate type was a syenite, with a CTE of 5.2 ×10-6/ºF (9.4 ×10-6/ºC).

The average PCC air content was 7.8 percent. Additional mixture properties are summarized in table 68 and table 69.

Table 68. Summary of PCC mix design information.
Variable Value
Cement type Type I
Cementitious material content (lb/yd3) 719
Water-to-Cement ratio 0.38
Curing method Membrane curing compound
1 lb/yd3 = 0.593 kg/m3.

 

Table 69. Summary of PCC pavement design features.
Variable Value
Joint spacing (ft) 15
Sealant type Liquid (silicone)
Dowel diameter (inches) 1.25
Dowel length (inches) 18
Dowel spacing (inches) 12
Edge support 10-ft HMA shoulder
Lane width (ft) 12
1 ft = 0.305 m.
1 inch = 25.4 mm.

 

Depth to Rigid Layer/Water Table.

The depth to the water table is also required by the MEPDG. However, no data were found regarding the location of the water table for test section 05-0218 in the LTPP Program database. Historical groundwater levels from the USGS Web site indicate an average of approximately 25 to 75 ft (7.6 to 22.9 m) below ground surface.(4) This monitoring location is not at the location of the 050218 section, but is located in Saline County, AR. A value of 25 ft (7.6 m) was adopted for the design of this section.

Climate/Environment Data.

Climate data were obtained from the updated climate files on the MEPDG Web site.(5) Sixteen climate stations were available in the MEPDG for the State of Arkansas. Considering both elevation and distance, the station in Little Rock was chosen for the study. This station was 52.9 mi (85.2 km) from the site and 18 ft (5.5 m) lower in elevation. The general weather category for the case study location is wet-nonfreeze.

Traffic Data.

AADT data were available from the LTPP Program database for 2 years: 1996 and 1998. The average AADTT data were available from the LTPP Program database for 2000, 2007, and 2008. These values are listed in table 70. Only the data from these years were used to calculate the compound growth factor. A linear regression of the traffic from these 3 years gave a compound growth factor of 1.9 percent and a two-way AADTT of 11,265 for 2009 with a R2 value of 0.99. Including traffic data from years 1996 and 1998 would decrease the R2 value to 0.14, so these values were omitted. Assuming a directional distribution factor of 0.5 and a design lane distribution factor of 0.95, 45 percent of the two-way truck traffic was in the design lane.

Table 70. AADTT from the LTPP Program database.

Year Two-Way AADTT
1996 7,360
1998 12,760
2000 9,684
2007 10,840
2008 11,080

 

ANALYSIS AND INTERPRETATION OF FWD TESTING DATA

This section presents the data checks; backcalculation analysis of the FWD data, LTE, and void detection, as well as a comparison of the backcalculation results with laboratory testing.

Preprocessing Deflection Data.

FWD testing was performed on five separate occasions—November 13, 1996; April 15, 1999; July 26, 2001; October 8, 2003; and September 16, 2004. The FWD data collected on September16, 2004, were selected to backcalculate the PCC elastic modulus and the k-value.

Table 71 is a summary of the deflection data for each location as well as the respective loads. The 9,000-lb (4,086 kg) load group was used for all backcalculation analysis. The data for the four drops at this load level were averaged and used for backcalculation. The loads and deflections used to determine joint load transfer characteristics and the presence of potential voids are shown in table 72 and table 73, respectively.

Table 71. Loads and deflections used to backcalculate the PCC elastic modulus and k‑value.
Station (ft) Location Drop load (lb) Deflection, mil
Sensor 9 Sensor 1 Sensor 2 Sensor 3 Sensor 4 Sensor 5 Sensor 6 Sensor 7 Sensor 8
35.10 J1 9,466 2.49 2.91 2.68 2.53 2.28 2.07 1.63 1.27 1.01
81.04 J1 9,248 3.35 3.62 3.35 3.14 2.80 2.47 1.85 1.34 0.93
126.97 J1 9,264 3.70 4.38 4.24 4.09 3.82 3.56 2.93 2.34 1.75
171.92 J1 9,201 2.86 3.62 3.61 3.50 3.23 2.94 2.14 1.38 0.92
216.86 J1 9,233 2.76 3.14 2.87 2.68 2.31 2.09 1.50 1.09 0.81
277.89 J1 9,148 2.95 3.24 3.01 2.83 2.61 2.35 1.90 1.48 1.13
323.16 J1 9,185 3.41 3.82 3.60 3.36 3.04 2.69 2.14 1.68 1.31
366.14 J1 9,169 3.14 2.93 2.55 2.36 2.07 1.93 1.57 1.29 1.05
413.06 J1 9,069 2.77 3.24 2.97 2.85 2.61 2.35 1.82 1.44 1.13
488.85 J1 9,116 3.27 3.73 3.57 3.44 3.19 2.94 2.47 2.01 1.46
1 ft = 0.305 m.
1 mil = 0.0254 mm.
1 lb = 0.454 kg.

 

Table 72. Loads and deflections used to determine the joint load transfer characteristics of test section 05-0218.

Station (ft) Joint Test Location Drop Load (lb) Deflection (mil)
Sensor 9 Sensor 1 Sensor 2
27.9 1 J4 9,143 3.74 5.04 4.08
28.9 1 J5 9,100 3.80 3.70 3.07
73.2 2 J4 9,127 2.56 3.30 3.15
74.2 2 J5 9,116 2.78 2.96 2.39
118.1 3 J4 9,111 3.26 3.88 3.44
119.1 3 J5 9,158 3.08 3.15 2.71
164.0 4 J4 9,233 2.55 3.28 2.80
165.0 4 J5 9,211 2.47 2.61 2.34
209.0 5 J4 9,095 2.44 2.78 2.54
210.0 5 J5 9,084 2.32 2.59 2.25
269.0 6 J4 9,164 3.93 4.57 4.22
270.0 6 J5 9,026 3.61 3.49 3.02
314.0 7 J4 9,105 3.89 4.56 4.34
315.0 7 J5 8,931 3.92 3.87 3.44
357.0 8 J4 8,973 3.80 4.69 4.44
357.9 8 J5 9,037 3.87 3.81 3.26
404.9 9 J4 8,947 3.52 4.02 3.54
405.8 9 J5 8,894 3.06 3.07 2.65
480.0 10 J4 8,962 3.13 3.64 3.47
481.0 10 J5 8,999 3.34 3.71 3.39
1 ft = 0.305 m.
1 mil = 0.025 mm.
1 lb = 0.454 kg.

 

Table 73. Loads and deflections used for void detection.
Station (ft) Location Drop Load (lb) Sensor 1 Deflection (mil)
8.8 J2 9,105 7.47
8.8 J2 12,204 9.54
8.8 J2 15,931 12.06
22.6 J2 9,133 5.60
22.6 J2 12,224 7.12
22.6 J2 16,097 8.84
36.3 J2 9,165 4.50
36.3 J2 12,188 5.93
36.3 J2 15,490 7.65
50.3 J2 9,272 3.87
50.3 J2 12,145 4.96
50.3 J2 15,593 6.29
64.0 J2 9,193 4.08
64.0 J2 12,252 5.59
64.0 J2 15,911 7.38
82.3 J2 9,113 4.21
82.3 J2 12,196 5.66
82.3 J2 16,101 7.43
96 J2 9,105 5.01
96.0 J2 12,109 6.58
96.0 J2 15,744 8.46
109.01 J2 8,903 5.77
109.1 J2 12,029 7.30
109.1 J2 15,891 8.76
123.7 J2 9,046 4.01
123.7 J2 12,117 5.39
123.7 J2 16,022 7.15
146.6 J2 9,058 5.29
146.6 J2 12,156 6.95
146.6 J2 15748 8.88
1 ft = 0.305 m.
1 mil = 0.025 mm.
1 lb = 0.454 kg.

 

The first step in the data analysis was to plot the deflection basin based on the measured deflections for each test location. The deflection basins for test section 05-0218 are shown in figure 73, where the legend indicates the test locations as presented in table 71. As shown in the figure, all of the locations showed good profiles except for the stations at 171.9 ft (52.4 m) and 366.1 ft (111.7 m). These locations had unacceptable profiles because of the irregular relationship between their load positions and deflections. The deflection data from these stations were not used in the backcalculation of the k-value and the PCC elastic modulus.

Click for description
1 mil = 0.0254 mm.
1 inch = 25.4 mm.

Figure 73. Graph. Deflection basins for test section 05-0218.

 

Secondly, to graphically evaluate the deflection variation along the section, the maximum deflections, presented in table 71, were normalized to the standard 9,000-lb (4,086 kg) load and plotted against distance, as shown in figure 74. As illustrated in figure 74, there was a variation in the structural response of the pavement along the section, particularly in the first approximately 150 ft (45 m), but there was generally no consistent trend.

Click for description
1 ft = 0.305 m.
1 mil = 0.0254 mm.

Figure 74. Graph. Normalized deflections along test section 05-0218.

 

Backcalculation Analysis.

Backcalculation of the test section 05-0218 deflection data was performed using the AREA60 method to determine the k-value and layer moduli. The analysis for this section consisted of a three-layer system: PCC slab, stabilized base, and spring foundation.

Backcalculation Results

Table 74 presents a summary of the determined AREA60 and radius of relative stiffness values for each location. The dynamic k-value and PCC elastic modulus for each station are presented in table 75 along with their overall static and dynamic average values.

Table 74. Summary of AREA60 and radius of relative stiffness.
Station (ft) Area60 Radius of Relative Stiffness (inches)
35.1 39.18 30.17
81.04 36.92 27.23
126.97 43.80 38.10
216.86 35.87 26.02
277.89 40.04 31.42
323.16 39.38 30.45
413.06 39.55 30.69
488.85 43.20 36.88
1 ft = 0.305 m.
1 inch = 25.4 mm.

 

Table 75. Average dynamic and static E and k-value.
Station (ft) Dynamic k-value (psi/inch) Dynamic PCC Elastic Modulus (psi)*
35.1 422 9,010,000
81.04 405 5,730,000
126.97 178 9,270,000
216.86 508 6,200,000
277.89 340 8,470,000
323.16 307 5,920,000
413.06 352 7,900,000
488.85 219 9,060,000
Average dynamic value 340 7,700,000
Average static value 170 6,180,000
*Composite elastic modulus of PCC slab and stabilized base.
1 ft = 0.305 m.
1 psi/inch = 0.263 kPa/mm.
1 psi = 6.89 kPa.

 

As illustrated in figure 75, there was variation in the k-value along the section, which was consistent with the variation observed in the normalized deflections. It appears that the k-value was extremely variable across the section. In general, a COV in the backcalculated k‑value of less than 20 percent, after screening of outliers, is reasonable.(10) Significantly higher k‑value COVs suggest significant changes in the subgrade soil type, the embankment thickness, or depth to bedrock. In this case, the COV of the backcalculated k-values was about 30percent, which suggests that a variation existed in the subgrade soil type, the embankment thickness, or the water table elevation along the section. Because of the lack of information about these parameters, it was not possible to determine a more precise cause of this variation.

Click for description
1 ft = 0.305 m.
1 psi/inch = 0.263 kPa/mm.

Figure 75. Graph. k-values along test section 05-0218.

 

The variation of the radius of relative stiffness along the section, which is the relative stiffness of the slab to that of the pavement foundation, is presented in figure 76. For this relationship, the COV was approximately 13 percent, which corresponded to the large COV found for the backcalculated PCC elastic moduli and the k-values along the section.

Click for description
1 inch = 25.4 mm.
1 ft = 0.305 m.

Figure 76. Graph. Radius of relative stiffness values along test section 05-0218.

 

The variation of the backcalculated PCC elastic modulus along the section is illustrated in figure 77. According to the backcalculated values, the COV for PCC elastic modulus was approximately 20 percent. This was higher than the 15 percent typically assumed to be acceptable for a PCC mixture design. In addition, the research team also found that, along the section, the high backcalculated k-values always correlated to low PCC elastic moduli. This can be an indication that the backcalculation process captured the overall stiffness, but it overestimated the k-value which in turn underestimated the elastic modulus of the PCC.

Click for description
1 psi = 6.89 kPa.
1 ft = 0.305 m.

Figure 77. Graph. PCC elastic modulus values along test section 05-0218.

 

Using the AREA method, an effective modulus of an equivalent single slab was first calculated. The method presented by Ioannides and Khazanovich, discussed in chapter 5, volume I, was then used to modify the backcalculated modulus for the PCC elastic modulus and to estimate a value for the base layer modulus.(15) An additional assumption must be made relating the modulus of the base layer to the modulus of the PCC layer (β) to obtain the separate moduli. Khazanovich, Tayabji, and Darter have presented a table with typical values for the β for different base materials.(16) Using this table, a value of 0.5 was selected for β for a lean concrete base layer. The layer bonding condition also must be determined; a bonded condition was assumed for this case study.

Using the backcalculated static modulus value of 6,180,000 psi (42,580,200 kPa) for Ee, an he of 7.8 inches (203 mm), a β value of 0.5, and assuming a bonded condition, E1 (PCC modulus) was calculated as 2.4 million psi (16,545,600 kPa) and E2 (cement-treated base) was 1.2 million psi (8,272,800 kPa). These were both much lower than expected for these layers. For this case, the backcalculated elastic modulus of the slab was modified from 6,180,000 to 2,400,000 psi (42,580,200 to 16,545,600 kPa), which meant that neglecting the stiffness of the stabilized base layer while backcalculating the PCC slab stiffness resulted in an overestimation of the modulus of the PCC slab.

On the other hand, comparing the results obtained for the PCC slab stiffness using the β-method 2.4 million psi (16,545,600 kPa)) with the ones measured in the laboratory in year 2006 3.6million psi (24,804,000 kPa)) implied that the β-method underestimated the PCC stiffness. Yet it should be noted that significant variability was observed in the PCC elastic modulus laboratory measurements. The significant drop of more than 1 million psi (6,894,000 kPa) in the modulus over 4 years can only be explained through test and measurements inconsistencies. For this reason, no judgments can be made on the modulus predicted for the base layer. However, the predicted value, 1.2 million psi (8,272,800 kPa), was lower than the default value used in the MEPDG, 2 million psi (13,788,000 kPa).

Furthermore, the selection of the β value can have a large effect on the results. Larger values of β mean a stiffer base layer and, therefore, more structural contribution to the effective modulus of the slab. When using this method, special care must be taken in selecting the β value.

Joint Load Transfer

Table 76 summarizes the calculated LTE and differential deflections based on the deflections presented in table 72. In table 76, all of the LTEs are above the acceptable level of 75 percent. In addition, all of the differential deflections were below 10 mil (0.25 mm), which supports the fact that the joints were performing well. The deflection data used for determining the LTEs were collected when pavement surface temperatures were 115 to 120 ºF (46.1 to 48.8 ºC). Such high surface temperatures increase the potential for joint lockup, so it was difficult to determine whether the joints were performing well at lower temperatures based on this testing.

Table 76. LTEs and differential deflections for the approach and leave slabs.
Joint Approach Slab LTE (percent) Approach Slab
Differential Deflection (mil)
Leave Slab
Differential Deflection (mil)
1 78 -0.002 0.000
2 92 -0.001 0.000
3 85 -0.001 0.000
4 85 -0.001 0.000
5 92 -0.013 -0.006
6 90 -0.001 0.000
7 94 -0.001 0.000
8 93 -0.001 0.000
9 87 -0.001 0.000
10 100 0.000 0.000
1 mil = 0.0254 mm.

 

Variation in LTE Over Time

Figure 78 shows the LTEs determined from the available data between 1996 and 2004. As shown in the figure, the LTEs varied considerably by year. The LTEs were lowest in 1996 and 1999 and highest in the remaining years. The variation was attributed to slab temperature at the time of testing, with the high LTEs likely the result of the extremely high temperatures locking up the joints. The pavement was constructed in 1993, and FWD testing was first performed in 1996 on a cloudy day with slab temperatures ranging from 45 to 50 ºF (7.2 to 10 ºC) at the time of testing. Even though the section was constructed with dowel bars, the LTE was poor for almost every transverse joint only 3 years after construction.

Click for description
1 ft = 0.305 m.
Figure 78. Graph. LTEs for different years at different testing locations.

 

In 1999, FWD testing was performed on a cloudy day with slab temperatures ranging from 70 to 75 ºF (21 to 24 ºC). LTEs in all of the joints were calculated to be unacceptable once again. FWD testing was performed in 2001 on a sunny day with slab temperatures ranging from 110 to 125 ºF (43 to 52 ºC). The LTEs for this day were all above the acceptable range, even though there had been no rehabilitation of the joints. Testing from 2003 indicated LTEs slightly lower than 2001. This testing was completed on a sunny day when the temperatures ranged from 95 to 100 ºF (35 to 38 ºC), which was slightly cooler than during the 2001 testing. Finally, FWD testing was completed in 2004 on a partly cloudy day when the temperature range was 115 to 120 ºF (46 to 49 ºC), and all of the joints again had an acceptable LTE. Therefore, it appears that the joints were not performing well except when the slab temperature was higher than 75 to 80 ºF (24 to 27 ºC) and the joints were locked.

Void Detection

The temperature distribution throughout the depth of the slab at the time FWD testing is performed can be critical when using the deflection data for void detection. If a positive temperature gradient is present in the slab, then the corners of the slab will curl downward so that erosion of the base might have occurred while not being detected. If a negative gradient is present, causing the slab corners to curl upward, erosion of the base might be detected even it has not occurred. To reduce the potential for false positive or false negative results, it is recommended that FWD testing be performed when the slabs are flat. The research team found that the slabs were most likely to be flat when a positive temperature gradient was present in the slabs because of the existence of built-in gradients.(12) As shown in figure 62 through figure 67 and in table 77, the average temperature gradient throughout the FWD test day was positive for all the slabs. The extent of these gradients, which were assumed to be linear, is presented in table 77.

Table 77. Slab temperatures and gradients for the FWD test performed on test section05‑0218.
FWD Test Date FWD Test Time Average Temperature Gradient for the Day of Testing (°F/inch) Average Slab
Temperature (°F)
11/14/1996 10:30 a.m. 0.14 50.8
04/15/1996 3:00 p.m. 0.95 70.1
04/19/1999 10:00 a.m. 0.89 64.2
07/26/2001 12:00 p.m. 1.96 99.5
10/08/2003 2:00 p.m. 2.39 88.8
09/16/2004 2:00 p.m. 3.42 101.9
1 °F/inch = 0.041 °C/cm.
°F = 1.8 × °C + 32.

 

Voids were determined using the void detection by maximum deflection method.(11) In this method, load versus deflection response for each station is determined. Voids were detected by determining the intercepts of each line of maximum deflections with the y-axis. If an intercept was greater than 2 mil (0.05 mm), there was potentially a void under the slab at that location. The intercept values from the load versus deflection plots for the FWD tests performed for test section 05-0218 are summarized in figure 79. As shown in this figure, the majority of the testing locations exhibited possible voids for most of the testing dates.

Click for description
1 mil = 0.0254 mm.
1 ft = 0.305 m.

Figure 79. Graph. Y-intercepts from void detection test results.

 

Based on the graph presented in figure 79, few voids were identified in 2001, 2003, and 2004, while voids could be detected at almost all joints in 1996 and 1999. This can be explained by the temperature gradients present in the slabs at the time of testing, as listed in table 77. According to these temperature measurements, the temperature gradient present at the time of testing was higher with each consecutive year. It appears that a negative built-in gradient might have been constructed into the slab. Void detection testing while the corners of the slab are curled upward can result in a false positive, as was observed in 1996 and 1999. As the gradient becomes increasingly positive at the time of testing, the slab corners curl downward until the slab is flat. It is possible for the slab corners to curl downward, with increasing positive gradients, to the point where the slab corners are penetrating into the lower layer. When this occurs, the y-intercept typically becomes negative, as observed in 2003 and 2004 when large positive gradients were present at the time of testing.

Backcalculation Modeling Issues and Recommendations.

The pavement structure for this section was a four-layer system or five-layer system if a rigid layer were present. The closed-form solutions for rigid pavements at the time of this report are not well suited for more than two bound layers and multiple unbound layers. As a result, the aggregate subbase layer is included in the determined k-value, as is the presence of any rigid layer. In this case, the aggregate subbase layer was thin, so the influence on the determined subgrade k-value should be minimal. Because it was a thin layer below a rigid layer, determining the layer modulus would be problematic with any backcalculation method, so combining it with the subgrade was an acceptable step for backcalculation. The available data indicated no near-surface rigid layer, so assuming any influence of a rigid layer in the composite k-value was acceptable for this example.

In addition, with the AREA method, an effective modulus for an equivalent single plate was first determined, and the individual bound layer moduli were then determined using a modular ratio. The determination of the separate moduli is influenced by the selection of the modular ratio and the assumed bonding condition. While general modular ratios are available, additional project-specific data assisting with establishing these parameters would be the best option.

Comparison of Backcalculation and Laboratory Testing Results.

To assist in evaluating which layer characteristics were appropriate to use in the MEPDG software, the results of the backcalculation (field tests) were compared with results obtained from laboratory testing conducted as part of the LTPP Program.

Bound Materials

As presented in table 75, the average backcalculated static PCC elastic modulus is 6,180,000 psi (42,718,000 kPa) whereas, according to the LTPP Program data, the laboratory measured static modulus for this section in 2006 is reported as 3.6 million psi (24,804,000 kPa). This difference is clearly significant and can be explained in two ways:

Unbound Materials

The granular subbase was a relatively thin layer directly under a stiff layer, which made it problematic to determine a modulus value from backcalculation. In addition, the influence of a thin unbound layer compared with the contribution of the underlying subgrade was minimal. Thus, the modulus of the subbase assumed the material’s default value of 206,820 kPa (30,000psi).

The resilient modulus of the subgrade ranged from approximately 9,000to 19,000 psi (62,046 to 130,986 kPa) for a bulk stress between 8 to 28 psi (55 to 193 kPa), as previously shown in
figure 68. Considering the fact that resilient modulus testing was not performed on any samples from test section 05-0218 and recognizing the difficulty in determining the resilient modulus for the unknown confinement condition, the typical resilient modulus suggested by the MEPDG for AASHTO A-2-4 subgrade, 16,500 psi (113,751 kPa), was adopted. The average backcalculated composite k-value corresponded to a modulus of approximately 19,000 psi (130,986kPa). This value corresponded with the laboratory values at the highest bulk stress, which was approximately 15 percent higher than the default material value.

RECOMMENDED REHABILITATION DESIGN INPUTS

The objective of this study was to assess the reliability of using FWD backcalculated values in design based on the comparison between designs with inputs from laboratory tests and designs with inputs obtained from FWD backcalculation. For this case study, rehabilitation designs were analyzed for an HMA overlay, an unbonded JPCP overlay, and a bonded JPCP overlay.

Inputs for the Existing Layers.

The general inputs used for the design program were based on the available LTPP Program data (as previously summarized), estimated inputs from standard specifications, and on default values within the program when data were not available from other sources. The primary inputs that were evaluated for this case study were those corresponding to properties obtained from the backcalculation process (primarily PCC elastic modulus and k-value).

Inputs for the HMA Overlay.

The overall design level for an HMA overlay over JPCP is level 3 and cannot be changed. The input level for the individual layers can be adjusted, with levels 1 through 3 for the HMA overlay and existing PCC, and levels 2 and 3 for unbound materials. For the HMA overlay analysis, the flexural strength and elastic modulus were entered for the existing PCC as required in level 1, and the unbound layer data were entered as level 3.

From the design runs performed during this case study, the HMA overlay design appeared to be very sensitive to the HMA properties of the overlay. The results presented later in this chapter are based on the following HMA mixture properties: the asphalt binder grade was selected as PG64-22, the effective binder content was 10 percent, and the air void content was 4 percent. The aggregate gradation is presented in table 78.

Table 78. Aggregate gradation for the new HMA overlay alternative design.

Sieve Size Value
3/4 inch (percent retained) 0
3/8 inch (percent retained) 18
No. 40 ( percent retained) 33
No. 200 (percent passing) 4
1 inch = 25.4 mm.

 

Table 79 summarizes the performance criteria used in the assessment of the HMA overlay design. These criteria are recommended by the MEPDG. A reliability level of 90 percent was assumed, as well as a 20-year performance period.

Table 79. Performance criteria for a HMA overlay on a rigid pavement.
Performance Criteria Limit
Initial IRI (inches/mi) 63
Terminal IRI (inches/mi) 172
Transverse cracking (percent slabs cracked) 15
AC surface down cracking (longitudinal cracking) (ft/mi) 2,000
AC bottom up cracking (alligator cracking) (percent) 25
AC thermal fracture (transverse cracking) (ft/mi) 1,000
Chemically stabilized layer (fatigue fracture) 25
Permanent deformation (AC only) (inches) 0.25
Permanent deformation (total pavement) (inches) 0.75
Reflective cracking (percent) 100
1 inch/mi = 0.0158 m/km.
1ft/mi = 0.19 m/km.
1 inch = 25.4 mm.

 

Inputs for the Unbonded JPCP Overlay.

The same PCC properties of the existing PCC layer were used for the JPCP unbonded overlay design. The properties based on laboratory measurements were as follows: unit weight of 138lb/ft3 (2.2g/cm3), Poisson’s ratio of 0.15, CTE of 5.2×10-6/°F (9.4 ×10-6/ºC), and an elastic modulus of 3.6 million psi (24,804,000 kPa). The PCC mixture design used for the existing PCC layer was also used for the overlay, such as a water-to-cement ratio of 0.38 and a cement content of 719 lb/ft3 (11.5 g/cm3). The existing PCC was not being rubblized or crack-and-seated, so the fracture type was left as “User Defined.” In addition, the existing PCC layer was fixed as a level3 input in an unbonded PCC overlay design.

The PCC slabs were designed to be 15 ft (4.6 m) long and 12 ft (3.7 m) wide to be compatible with the existing PCC layer, and the joints were sealed with silicon and doweled with 1.25-inch (31.8-mm) bars. With respect to the required bond breaker layer (2 inches (50.8 mm) thick, as suggested by the MEPDG), the same HMA properties were employed as were used in the HMA overlay design alternative.

The performance criteria for the unbonded JPCP overlay suggested by the MEPDG are presented in table 80. A 90-percent reliability level and 20-year performance period were used.

Table 80. Performance criteria for a PCC overlay on a rigid pavement.
Performance Criteria Limit
Initial IRI (inches/mi) 63
Terminal IRI (inches/mi) 172
Transverse cracking (percent slabs cracked) 15
Mean joint faulting (inches) 0.12
1 inch/mi = 0.0158 m/km.
1 inch = 25.4 mm.

 

Inputs for the Bonded JPCP Overlay.

The inputs, as well as the performance criteria for the JPCP bonded overlay, were the same as those used for the unbonded JPCP overlay with the difference that the bonded overlay did not include an asphalt interlayer. In addition, the input level for the existing PCC layer can be based on levels 1 through 3 for a bonded PCC overlay.

REHABILITATION DESIGN RESULTS

Three types of overlays were designed as the rehabilitation for the existing JPCP, namely an HMA overlay, an unbonded JPCP overlay, and a bonded JPCP overlay. To validate the reliability of using FWD backcalculation results in MEPDG designs, two design alternatives were originally planned for each kind of overlay, the first of which would employ the laboratory testing results as inputs, and the second would employ the backcalculated values as inputs. However, the research team later found (from a study on the conversion of layer moduli into the k-value, which is part of this report) that the MEPDG software did not appear to always use the entered dynamic k-value to represent the stiffness of all the layers beneath the base layer as is stated in the MEPDG documentation. Sometimes, the stiffness of the base layer is also taken into account in the composite k-value. Therefore, a third alternative had to be designed to learn how the MEPDG uses the input k-value. The details of the following three design alternatives are presented:

HMA Overlay of JPCP.

In the MEPDG HMA rehabilitation design, the thickness of the overlay was varied until the thinnest HMA layer was obtained that still satisfied the selected performance criteria. The HMA overlay thickness determined to meet all criteria at the 90-percent reliability level with the assumed inputs was 15 inches (381 mm), as illustrated in figure 80, regardless of the different sets of inputs described for alternatives 1 through 3. The rutting criterion is what drives this excessively thick HMA overlay, which is considered impractical from a constructability standpoint.

Click for description
1 inch = 25.4 mm.
Figure 80. Graph. Summary of predicted reliabilities for surface rutting of HMA overlay.

 

Thus, in this case study, the structural capacity of the existing PCC pavement did not control the design, but rather it was controlled by the HMA overlay material properties. Therefore, the final HMA overlay thickness requirement, regardless of the underlying layer properties selection, the same. This implies that changes in the HMA mix design would substantially affect the design thickness of the HMA overlay. However, several variations in mix properties were analyzed (such as varying binder grades, binder content, air void content, etc.), and rutting continued to control the design results. If rutting was not considered, an HMA overlay thickness of 4 inches (102 mm) was obtained that successfully meets the other performance criteria.

The research team also found that the k-values reported in the output file were the same whether or not the dynamic k-value was used as an input. The reported k-value appears to be based on the entered subgrade layer modulus, which was confirmed by running different moduli with and without the dynamic k-value entered. Additional design runs were executed with low and high base layer moduli to determine how the k-value was calculated. The k-values were identical for cases with varying base layer stiffness, indicating the stiffness of the base was not included in the k-value calculation.

Design Results for the Unbonded PCC Overlay.

The design thickness of the unbonded JPCP overlay was the thinnest slab that still satisfied the selected performance criteria (shown in table 80). However, the use of unbonded JPCP overlays thinner than 7 inches (178 mm) is not recommended by the MEPDG. This is because the unbonded overlay acts independently from the supporting layers, so its minimum allowable thickness should be limited to ensure its structural capacity as well as its functionality. As summarized in table 81, the research team found that a 7-inch-(178-mm)-thick or thicker JPCP unbounded overlay can meet all of the criteria for the first two alternatives; however, because the elastic modulus of the PCC estimated in alternative 3 was very low, the structure failed in transverse cracking and requires a greater thickness. In that case, a 15-inch (381-mm) PCC overlay would be required to satisfy the selected performance criteria.

Table 81. Design thickness of unbonded PCC overlay for three design alternatives.

Alternative Design Thickness (inches)
1 7
2 7
3 7
1 inch = 25.4 mm.

 

Additional design runs for unbonded PCC overlays were made to assess the k‑value to determine whether it was computed similarly as was observed in the HMA overlay assessment. Based on the additional design runs, the mean of the calculated k-values agreed well with the entered k‑value suggesting that the MEPDG uses the manually entered k-value for unbonded JPCP overlay designs. However, when no k-value was designated, it was not explicit which layers were taken into account in the calculation of the k-value. However, a noticeable difference was found between using a low and high existing PCC modulus, which might indicate that the stiffness of the existing PCC was involved in the calculation of the k-value.

Design Results for the Bonded PCC Overlay of JPCP.

The design thickness of the bonded JPCP overlay was determined by reducing the thickness of the overlay until the performance criteria shown in table 800 were no longer fulfilled. As presented in table 82, the design thickness for all three alternatives is 3 inches (76 mm). Thinner slabs were not recommended by the MEPDG version used in the analysis (Note that the newer version 1.1 of the MEPDG allows a 1.5-inch minimum).

Table 82. Design thickness of bonded PCC overlay for three design alternatives.

Alternative Design Thickness (inches)
1 3
2 3
3 3
1 inch = 25.4 mm.

 

The research team found that the correlations between the elastic modulus and modulus of rupture of PCC used by the MEPDG were different for bonded and unbonded overlays. In other words, using the same modulus of rupture for bonded and unbonded cases resulted in different elastic moduli.

As with the HMA and unbonded PCC overlay alternatives, additional design runs were executed to determine the k-value calculations. The mean k-value for a high and low modulus base material noticeably varied. Therefore, it seems that the modulus of the base layer was considered in the calculation of the k-value for bonded JPCP overlay designs, which agreed with the assumptions made for bonded JPCP overlays. It is also apparent that the calculated k-values matched the entered dynamic k-value.

Evaluation of Design Results.

An HMA overlay was designed for rehabilitation of test section 05-0218 in three ways. First, measured laboratory moduli of the pavement layers were used for the design. Second, using available FWD test results, the elastic modulus of PCC and dynamic effective k-value were backcalculated and used for the design. Finally, the backcalculated PCC elastic modulus was modified and reduced to account for the contribution of the stabilized base layer. The required thickness was found to be the same for the three alternatives. This is attributed to the HMA overlay requirement being controlled by HMA materials performance and not the underlying structural capacity in this case. In addition, the research team found that it did not appear the dynamic k-value was being used when it was used as an input.

An unbonded PCC overlay design was also analyzed using the same three alternatives. A 7-inch (178‑mm) PCC overlay met the selected performance criteria over the 20-year design life for alternatives 1 (laboratory-based inputs) and 2 (backcalculated PCC modulus and k-value). However, alternative 3 (modified PCC elastic modulus and base modulus) required a much thicker unbonded overlay (15 inches (381 mm)). The modified elastic modulus value was as low as 2.4 million psi (16,536,000 kPa), which did not provide enough structural capacity to adequately carry the future traffic loadings.

The final rehabilitation design method used was a PCC bonded overlay. A 3-inch (76-mm) bonded overlay was determined to be sufficient to meet the selected design criteria for all threedesign alternatives. A 76-mm (3-inch) PCC overlay met the design criteria even in the third alternative, which included a lower strength PCC, which can be explained by the fact that the existing pavement and the overlay performed as a monolithic section. In general, before the bonded overlay is placed on top of an existing JPCP, all existing distresses are rehabilitated and a good bond is achieved between the new and the existing pavement. Therefore, in the case of alternative 3, the new overlay (3 inches (76 mm)) and the existing pavement (7.8inches (198mm)) combined to produce a composite PCC layer placed on a strong base layer (lean concrete with modulus of 2 million psi (13,788,000 kPa). This section proved to have sufficient structural capacity to carry the loads over the design life with a reliability of 90percent.

For unbonded overlays, the existing slab acts a base layer. The low modulus defined for the overlay in the case of alternative 3 (2.4 million psi (16,536,000 kPa)) required the slab to have a relatively higher thickness to provide sufficient structural capacity to bear the cumulating fatigue damage. Although the existing PCC slab with a modulus of 2.4 millionpsi (16,536,000 kPa) provided a strong base layer for the slab, numerous runs of the MEPDG revealed that a 15-inch (381-mm) overlay would be required to meet the transverse cracking criteria of 15percent with a 90-percent reliability.

SUMMARY

Test section 05-0218, located in Saline County, AR, was selected for rehabilitation case study of a PCC slab on stabilized base layer. The distress survey data available for this section in the LTPP Program database indicated that this section was in poor condition (50 percent of the slabs had working transverse cracking) in 2007.

The most recent FWD test data available for test section 05-0218 was from 2004. Using the LTPP Program database, input parameters required for the overlay designs were compiled. Not all required data were available for test section 05-0218; therefore, data from other sections of project 05-0200 were used. Also, the USGS Web site was used as a reference for data concerning the depth of the water table.

Using the FWD test data, the elastic modulus of PCC and the dynamic effective k-value were backcalculated for the section. These values were calculated to be 6,180,000psi (42,580,200kPa) and 340 psi/inch (100 kPa/mm), respectively. The backcalculated PCC elastic modulus was later corrected for the effect of the stabilized base layer. The corrected value of the PCC elastic modulus obtained was 2.4 million psi (16,536,000 kPa), which was lower than the measured value of 3.6 million psi (24,804,000 kPa). The lower PCC modulus appears to influence only the required thickness of the unbonded PCC overlay for this case study, as summarized in table 81. With the exception of the unbonded PCC overlay alternative, there was no apparent difference between the use of laboratory-based and backcalculation-based inputs based on the results of the rehabilitation designs.

Other observations made during the design analysis process include the following:

Federal Highway Administration | 1200 New Jersey Avenue, SE | Washington, DC 20590 | 202-366-4000
Turner-Fairbank Highway Research Center | 6300 Georgetown Pike | McLean, VA | 22101