Using Falling Weight Deflectometer Data With Mechanistic-Empirical Design and Analysis, Volume I: Final Report

CHAPTER 7. USING FWD DATA IN THE MEPDG—CASE STUDIES

INTRODUCTION

As part of this project, case studies featuring existing pavement sections were conducted to showcase the use of FWD data for the characterization of pavement layer properties for analysis with the MEPDG. The case studies demonstrate the use of FWD and laboratory testing in deriving strength-related input properties for the MEPDG and also allow comparisons of the input properties derived from the FWD and laboratory testing. These case studies help outline the pros and cons of laboratory versus FWD testing to develop strength-related input properties of in-service pavement layers for use in the MEPDG rehabilitation design.⁽⁷⁾

Six case study projects were evaluated, representing the following six pavement types:

• Flexible pavement.

• Flexible pavement on rubblized PCC.

• Rigid pavement on a granular base.

• Rigid pavement on a stabilized base.

• Rigid pavement on an existing HMA pavement.

• Composite pavement (flexible pavement overlay on rigid pavement).

Although several data sources were considered (i.e., MnROAD, LTPP Program, National Center for Asphalt Technology (NCAT), and sections of roadways that have been used for special studies by State highway agencies), ultimately, data from existing LTPP pavement test sections were used in the case studies.

AVAILABLE CASE STUDY DATA—LTPP DATABASE

To accurately assess how well FWD testing results can estimate the true in situ material properties for different pavement types, backcalculated material properties needed to be compared with material properties determined by more traditional field and laboratory testing procedures. Two basic approaches for accomplishing this task were to (1) find existing data sources (databases) or past special study projects that contain both material testing and FWD results, or (2) conduct sampling and testing at new project locations to collect data that can be analyzed under this study. Because six different pavement types are included in this study, a very large research effort would be required to conduct both FWD testing and field and laboratory testing at multiple projects. Consequently, known existing data sources, including the LTPP database, the MnROAD test sections, NCAT test sections, and various special State studies, were evaluated for possible use in this study. Ultimately, it was determined that the LTPP database provided the most complete and comprehensive data source for this project, and appropriate sections were identified to fit within the established pavement categories.

Specifically, the sites in the Specific Pavement Studies (SPS)-1, -2, -5, and -6 experimental sections in the LTPP Program were reviewed for use as case study projects. Along with new flexible and rigid pavement designs (SPS-1 and -2), these sections also contained rehabilitated HMA and PCC pavements (SPS-5 and -6). Specific descriptions for each of the aforementioned SPS categories (presented in the following sections) are reproduced here verbatim from the Long-Term Pavement Performance Information Management System: Pavement Performance Database User Reference Guide:⁽¹¹⁴⁾

SPS-1: Structural Factors for Flexible Pavements. The experiment on the structural factors for flexible pavements study (SPS-1) examines the performance of specific AC-surfaced pavement structural factors under different environmental conditions. Pavements within SPS-1 must start with the original construction of the entire pavement structure or removal and complete reconstruction of an existing pavement. The pavement structural factors in this experiment include the in-pavement drainage layer, surface thickness, base type, and base thickness. The experiment design stipulates a traffic loading level in the study lane in excess of 100,000 80-kN (18-kip) [equivalent single axle loads] ESALs per year. The combination of the study factors in this experiment results in 24 different pavement structures. The experiment is designed using a fractional factorial approach to enhance implementation practicality, permitting the construction of 12 test sections at one site and a complementary 12 test sections to be constructed at another site within the same climatic region on a similar subgrade type.

SPS-2: Structural Factors for Rigid Pavements. The experiment on the structural factors for rigid pavements study (SPS-2) examines the performance of specific JPCP structural factors under different environmental conditions. Pavements within SPS-2 must start with the original construction of the entire pavement structure or removal and complete reconstruction of an existing pavement. The pavement structural factors included in this experiment are in-pavement drainage layer (edgedrains or no edgedrains), PCC surface thickness (254 to 279 mm [10 or 11 in]), base type (dense-graded untreated granular, lean concrete, and permeable asphalt treated), PCC flexural strength (3.8 or 6.2 MPa [550 to 900 lb/in²]), and lane width (3.66 and 4.27 m [12 to 14 ft]). The experiment requires that all test sections be constructed with perpendicular doweled joints at 4.9-m (15-ft) spacing and stipulate a traffic loading level in the lane in excess of 200,000 ESALs per year. The experiment is designed using a fractional factorial approach to enhance implementation practicality, permitting the construction of 12 test sections at one site and a complementary 12 test sections to be constructed at another site within the same climatic region on a similar subgrade type.

SPS-5: Rehabilitation of Asphalt Concrete Pavements. The experiment on the rehabilitation of HMA pavements (SPS-5) examines the performance of eight combinations of HMA overlays on existing HMA-surfaced pavements. The rehabilitation treatment factors included in the study are the intensity of surface preparation, recycled versus virgin HMA overlay mixture, and overlay thickness. The experiment design includes all four climatic regions and the condition of the existing pavement. The experiment design stipulates a traffic loading level in the study lane in excess of 100,000 80-kN (18-kip) ESALs per year.

SPS-6: Rehabilitation of Jointed Portland Cement Concrete Pavements (JPCP). The experiment on the rehabilitation of JPCP pavements (SPS-6) examines the performance of seven rehabilitation treatment options as a function of the climatic region, type of pavement (plain or reinforced), and the condition of the existing pavement (note: an eighth scenario looks at just applying routine maintenance). The rehabilitation methods include surface preparation (limited preparation or full concrete pavement restoration) with a 102 mm- (4 in-) thick HMA overlay or without an overlay, crack/break and seat with two HMA overlay thicknesses (102 or 203 mm [4 or 8 in]), and limited surface preparation with a 102 mm- (4 in-) thick HMA overlay with sawed and sealed joints. (p. 142–143)

An assessment of the LTPP data availability using the LTPP DataPave (both online and disk-based) interface revealed 18 SPS-1 sections, 14 SPS-2 sections, 18 SPS-5 sections, and 14 SPS-6 sections that met all of the outlined LTPP criteria. In addition to those documented sections, the LTPP database also contained many “supplemental” sections that were also included in the LTPP database. Supplemental sections were typically State transportation department experimental projects that focused on investigating only one or two of the variables rather than all of the specific variables outlined in the LTPP project. The LTPP data also included key material testing results, performance monitoring data, climatic information, traffic loading data, and seasonal testing information.

CASE STUDY SELECTION

Although a number of different potential sections were considered for the case studies, it was determined that a relatively small number of sections contained data for all of the input information required in the MEPDG. Even those ultimately selected for the case studies (which were those having the most complete data) often did not have all of the desired data.

This section describes the pavement projects that were selected for the case studies, as summarized in table 13. The detailed case studies themselves are included in volume II of this report.

Table 13. Summary of selected case study pavement sections.

Case Study 1: Flexible Pavement

As discussed in chapter 4, the required input material properties for HMA pavements in the MEPDG relevant to the use of FWD data and backcalculation results are the following:

• Time-temperature dependent dynamic modulus E* for the HMA layer(s).

• Resilient moduli for the base/subbase and subgrade materials.

• Elastic modulus of the bedrock, if present.

The LTPP database contained 18 applicable SPS-1 sections representing all 4 climatic zones in the United States (i.e., wet-freeze, wet-nonfreeze, dry-freeze, and dry-nonfreeze). After reviewing the preliminary flexible pavement sections, LTPP section 30-0100 (30-0113 specifically) was selected as a flexible pavement rehabilitation case study. The test section, located on I-15 near Great Falls, MT, is a flexible pavement cross section with a HMA surface and aggregate base layer over subgrade. The original pavement, a 102-mm (4-inch) HMA surface on a 203-mm (8-inch) aggregate base, was constructed in 1997. Rehabilitation and repair work was performed in 2003 and 2004 (construction numbers 2 and 3, respectively).

This section was representative of the following selection factors:

• Dry-freeze climate zone.

• Principal interstate—rural functional class.

• “Fair” pavement condition.

• Sand subgrade classification.

• No (or deep) rigid layer.

Case Study 2: Flexible Pavement on Rubblized PCC

The MEPDG analysis of a flexible HMA overlay placed on a rubblized PCC pavement is very similar to the analysis of an HMA pavement placed on an aggregate base. The required MEPDG input material properties for this pavement type that are relevant to the use of FWD data and backcalculation results are the following:

• Time-temperature dependent dynamic modulus E* for the HMA overlay layer.

• Elastic modulus of the rubblized PCC layer.

• Resilient moduli for the base/subbase and subgrade materials.

• Elastic modulus of the bedrock, if present.

There were 17 rubblized test sections (representing 7 States) included in the LTPP database as SPS-6 sections. Section 17-0600 was selected for the flexible pavement on rubblized PCC case study. Section 17-600, located on I-57 near Pesotum, IL, consists of 14 test sections, with test section 17-0663 being selected for use in the case study.

The test section 17-0663 PCC pavement was originally constructed in 1964 (construction number 1), with rehabilitation/repair work performed in 1990 (May and June) and 1997 (construction numbers 2, 3, and 4, respectively). Based on LTPP core data, the average original pavement cross section consisted of 254 mm (10 inches) of PCC (jointed reinforced concrete pavement (JRCP)) and 178 mm (7 inches) of aggregate base. The JRCP was rubblized and a HMA overlay was placed in 1990.

This section was representative of the following selection factors:

• Wet-freeze climate zone.

• Rural principal arterial–interstate functional class.

• “Fair” pavement condition.

• Fine-grained soil subgrade classification.

• No (or deep) rigid layer.

Case Study 3: Rigid Pavement on Granular Base

When using the MEPDG procedure to develop an overlay design for an existing rigid pavement on a granular base, the following four-layer strength-related properties are required:

• Elastic modulus of the existing PCC layer.

• Resilient modulus of the existing base layer(s).

• Subgrade k-value.

• PCC flexural strength.

The preliminary review of data sources indicated that such data were readily available in the LTPP database. For this pavement type, the SPS-2 database contained 14 applicable projects. Each of these SPS-2 projects contained four different test sections with aggregate bases. (Note that the four test sections represented unique combinations of pavement thickness, PCC flexural strength, and lane width.)

LTPP test section 32-0200 was selected for the rigid pavement on granular base case study. The test section was a part of I-80 in Lander County, NV. According to the LTPP data, the JPCP consists of a 295-mm (11.6-inch) PCC surface layer, a 145-mm (5.7-inch) dense graded aggregate base, a 513-mm (20.2-inch) granular subbase, and a subgrade of which the top 305mm (12 inches) were treated with lime.

This section was representative of the following selection factors:

• Dry-freeze climate zone.

• Principal Interstate–Rural functional class.

• “Poor” pavement condition.

• Sandy silt subgrade classification.

• No (or deep) rigid layer.

Case Study 4: Rigid Pavement on Stabilized Base

When using the MEPDG procedure to design a rigid pavement on a stabilized base, the same layer strength-related properties required for the rigid pavement on a granular base pavement type are required. Those properties include the following:

• Elastic modulus of the existing PCC layer.

• Elastic modulus of the existing base layer(s).

• Subgrade k-value.

• PCC flexural strength.

Data required to verify the recommended FWD guidelines for this pavement type were also readily available from the SPS-2. For this pavement type, SPS-2 database contained 14different projects, each of which contained 4 different test sections with a lean concrete base, and 4 test sections with a permeable asphalt treated base. Similar to the PCC pavement on a granular base, the four different test sections associated with each stabilized base type represented unique combinations of pavement thickness, PCC flexural strength, and lane width.

LTPP project 05-0200, located on I-30 in Saline County, AR, was selected for the rigid pavement on stabilized base case study. The section consisted of 12 test sections, with test section 05-0218 selected for the case study. The PCC pavement was originally constructed in September 1993. Based on LTPP core data, the typical cross section of section 05-0218 consisted of 203 mm (8inches) of PCC (surface layer), 178 mm (7 inches) of a treated base (lean concrete), 102 mm (4inches) of a granular layer, a woven geotextile interlayer, and the subgrade.

Section 05-0218 was a rigid pavement representative of the following selection factors:

• Wet-nonfreeze climate zone.

• Rural principal arterial–interstate functional class.

• “Poor” pavement condition.

• Gravel subgrade classification.

Case Study 5: Rigid Pavement on HMA Pavement

The design of a rigid pavement on top of an existing HMA pavement (i.e., “whitetopping”) is equivalent to designing a PCC pavement with an HMA base, so the inputs required by the MEPDG are similar to those required by the other rigid pavement types. Specifically, the layer strength-related inputs required in the MEPDG approach include the following:

• Elastic modulus of the existing HMA layer

• Elastic modulus of the existing base layer(s)

• Subgrade k-value.

Data required for the rigid pavement overlay on an HMA pavement were obtained from the test section (30-0113) used in case study 1. Specific details on that pavement and data were provided in the previous description of case study 1.

Case Study 6: Composite Pavement

A composite pavement is defined as a PCC pavement that is overlaid with one or more HMA overlays. When analyzing composite pavements using the MEPDG procedure, the same input values required by the previously discussed cases studies were also required here.⁽⁷⁾ Specifically, the required inputs for such an analysis were the following:

• Elastic modulus of the existing PCC layer.

• Flexural strength of the existing PCC layer.

• Resilient moduli for the base/subbase and subgrade materials.

Although the load bearing of an HMA/PCC pavement is dominated by the underlying PCC pavement, the evaluation of FWD data collected on HMA-overlaid PCC pavements warrants special considerations because of the compression of the HMA layer. Therefore, for this pavement type, it was important to find case study projects in which the materials properties and construction procedures were well documented.

Fourteen SPS-6 projects were included in the LTPP database. As mentioned previously, each SPS‑6 project was used to investigate eight different rehabilitation options for JPCP pavements, three of which involved applying HMA overlays after completing various levels of pre-overlay repair. Specifically, the three rehabilitation options deemed useful to this study were the following:⁽¹¹⁵⁾

• SPS-6 section 03: Minimum restoration followed by a 100-mm (4-inch) HMA overlay. (Minimum restoration typically consists of routine maintenance, which includes limited patching (filling potholes), crack repair and sealing, and stabilization of joints.)

• SPS-6 section 04: Maximum restoration followed by a 100-mm (4-inch) HMA overlay with sawed and sealed joints. (Maximum restoration included activities such as subsealing, subdrainage, joint repair, full-depth repairs with restoration of load transfer, and shoulder rehabilitation.)

• SPS-6 section 06: Maximum restoration followed by a 100-mm (4-inch) HMA overlay.

In addition to the general SPS-6 sections, 13 additional supplemental SPS-6 sections representing 9 States were also available in the LTPP database.

Project 19-0600 from the LTPP SPS-6 database served as the basis for the composite pavement case study. The 19-0600 project was located on I-35 near Des Moines, IA, and test section 19‑0659 was selected as a composite pavement rehabilitation case study. The test section was a composite pavement cross section with a HMA surface (overlay) on a JRCP and granular base layer over subgrade. The original JRCP pavement, a 254-mm (10-inch) PCC slab on a 203-mm (8-inch) base with 23.3-m (76.5-ft) joint spacing, was constructed about 1965. A 102-mm (4‑inch) HMA overlay was placed in 1989.

This section was representative of the following selection factors:

• Wet-freeze climate zone.

• Principal arterial–Interstate (rural) functional class.

• “Fair” pavement condition.

• Clay subgrade classification.

• No (or deep) rigid layer.

SUMMARY OF CASE STUDY RESULTS

The selected project sites were used to backcalculate layer properties from deflection data and to develop design inputs in the MEPDG design procedure. This was done to illustrate the use of FWD-based data in the MEPDG and to evaluate the possible differences in results between FWD-based inputs and laboratory (and/or default) design inputs. Brief summaries of the significant findings from each case study are provided in the following sections, with details of the case studies presented in volume II of this report.

For the flexible backcalculation analyses, the following three backcalculation programs were used: MODTAG, MICHBACK©, and EVERCALC©. These programs were selected because they were widely used, readily available, and not proprietary. Three programs were employed to look into the effect of different inverse routines on backcalculated parameters and ultimately on rehabilitation design. For the rigid pavement backcalculation, the AREA method and outer-AREA method (for HMA/PCC pavements) were used because these had an established basis for use and the closed-form equations were easily implemented in a spreadsheet.

Case Study 1: Flexible Pavement

Three backcalculation programs (MODTAG, MICHBACK©, and EVERCALC©) were used to analyze FWD deflection data from various test locations (stations) within the project. The deflection data showed considerable variability within the project. In addition, the RMS values obtained from the three backcalculation programs were generally very high. It is recommended that agencies conduct FWD tests at multiple locations and use the average of backcalculated layer moduli for the section. This should provide consistent service life predictions irrespective of the backcalculation program used and therefore result in a similar recommended overlay design.

For this case study, the MEPDG results indicated that surface-down cracking was critical in the rehabilitation design of an HMA overlay over existing HMA pavements. Nevertheless, a 76-mm (3-inch) HMA overlay was satisfactory for nearly all of the input combinations. Within the ranges identified, the selection of inputs was more critical as one approaches the lower values for any layer.

In addition, the following recommendations are made:

• Until ongoing studies are completed, the correction factors for backcalculated properties of laboratory values developed by Von Quintus and Killingsworth should be applied to the backcalculation results.^(87,116)

• While the procedure needs to be verified, for the time being, an equivalent frequency of 30 Hz should be used. This is calculated as 1/(FWD pulse duration) = 1/0.033s = 30 Hz. Although this formula is technically incorrect, it is compatible with the equivalent frequency used for calculating E* to be used in MEPDG.

Case Study 2: Flexible Pavement on Rubblized PCC

The same three backcalculation programs used for the flexible pavement case study were used to analyze FWD deflection data from various test locations (stations) within the project. The deflection data showed considerable variability within the project. In addition, the RMS values obtained from the MICHBACK© backcalculation program were very high. As with the previous case study, it is recommended that agencies conduct FWD tests at multiple locations and use the average of backcalculated layer moduli. This should provide consistent service life predictions irrespective of the backcalculation program used, and therefore result in a similar recommended overlay design.

For HMA overlays on rubblized concrete pavements, the critical performance measure in the MEPDG software was surface rutting. The required HMA overlay thickness to achieve a 90‑percent reliability level was 178 mm (7 inches) for all analyses. However, if surface rutting was addressed through maintenance at some intermediate year (less than 20 years), a thinner HMA overlay would be appropriate.

The surface rutting predictive model was mainly sensitive to HMA overlay thickness and HMA and rubblized layer moduli. Therefore, care should be taken in selecting the modulus for the rubblized layer. In this particular case, the rubblized PCC modulus was slightly lower than the suggested level 3 input in the MEPDG. It may be useful to combine the rubblized layer with the existing unbound base layer when using some backcalculation programs (for example MICHBACK©, in this case). The remaining unbound layer moduli should be entered as the adjusted backcalculated values (until ongoing studies are completed, the correction factors developed by Von Quintus and Killingsworth should be applied to the backcalculation results), but the existing HMA modulus determined by backcalculation cannot currently be used.⁽¹¹⁵⁾

Case Study 3: Rigid Pavement on Granular Base

The backcalculation of PCC elastic modulus and modulus of subgrade reaction was only possible for the first half of section 32-0203 (station 0 to 75.6 m (0 to 248 ft)) because of the inconsistency in the deflection basin profiles for the rest of the stations. However, the information included in the LTPP database for this section was sufficient to determine the load transfer characteristics and the support conditions for the entire section.

The variation of the backcalculated k-value and the PCC elastic moduli along the section can be an indication that the backcalculation technique was capturing the overall stiffness, but it appeared to be overestimating the k-value and, therefore, underestimating the elastic modulus of the concrete for several of the test locations. The average laboratory-measured static PCC modulus 19,292,000 kPa (2.8 million lb/inch²) showed a good correlation with the average backcalculated static PCC elastic modulus of 21,359,000 kPa (3.1 million lb/inch²); however, the variation of the backcalculated values along the section was 22 percent, which was substantially higher than the 15percent typically assumed acceptable.

The composite k-value for the section was obtained based on the modulus that was determined for each layer and compared with the composite k-value from the backcalculation. The resulting static composite k-value was approximately 80 kPa/mm (300 lb/inch²/inch) for a base with 145‑mm (5.7-inch) thickness, which was much higher than the corresponding static composite k‑value obtained from the backcalculation (50 kPa/mm (190 lb/inch²/inch). This was in keeping with other observations that the backcalculated modulus of the unbound materials tended to be smaller than the laboratory-determined modulus values.

The LTE values along the section were above a minimum acceptable level of 75 percent. This high level was constant over time, and it was not affected by cold temperatures, indicating the doweled joints were performing well. According to the void detection analysis, it appears that voids were not present beneath the slabs in this section of roadway. However, there was a possibility that some erosion began to develop in 2002.

The studied section was in poor condition, with a large number of transverse cracks. Not all of the necessary MEPDG design inputs were available in the LTPP database for the studied section 32-0203, or any other 32-0200 sections.⁽⁷⁾ Therefore, based on the LTPP data, appropriate inputs for rehabilitation designs in the MEPDG are discussed in detail in the case study.

Three different overlay rehabilitation designs were developed for this project: an HMA overlay, an unbonded JPCP overlay, and a bonded JPCP overlay. The thinnest design thickness (102 mm (4 inches)) was obtained by using a bonded JPCP overlay because it used the remaining structural capacity of the existing road. The design thickness of the unbonded overlay (178 mm (7 inches)) was determined considering the fact that the unbonded JPCP overlays worked independently and thus some restraints in thickness must be provided to guarantee its structural capacity. The MEPDG provided an unreasonably thick HMA overlay design (305 mm (12inches)), with the critical performance parameter being surface rutting. Even when various modifications were made to the HMA mix design, the resulting HMA overlay thickness was considered unreasonably thick. If surface rutting were addressed through maintenance at some intermediate year (less than 20 years), then a thinner HMA overlay would be appropriate.

No difference in the design thickness was found among the three design alternatives (laboratory/ material default values, adjusted backcalculated k-value and backcalculated PCC elastic modulus, and backcalculated PCC elastic modulus and k-value), indicating the reliability of using both the backcalculated dynamic elastic modulus for the PCC layer and the dynamic k‑value for the supporting layers in the MEPDG design.

The backcalculated k-value that represents the composite stiffness of all layers beneath the slab can be directly entered into the MEPDG without having a significant influence on the design thickness for the pavement structure analyzed. However, this does not definitively mean that the MEPDG takes the stiffness of the base layer into account in the k-value. It could be either due to the insensitivity of the design thickness on the input k-value or because the granular base contributes little to the composite stiffness of all layers beneath the slab. Other observations made regarding the calculation of k-value within the MEPDG program included the following:

• For the HMA overlay design, varying the subbase stiffness had very little influence on the determined k-value, which seemed to suggest the stiffness of the subbase layer was not taken into account in the calculated k-value. In addition, the reported k-values in the output file were identical for cases with varying base layer stiffness, indicating the stiffness of the base was likely not included in the k-value calculation. Furthermore, it appears that the MEPDG ignored the entered dynamic k-value and calculated the k-values based on the entered layer moduli because the summarized values were the same regardless of what dynamic k-value was entered or when a dynamic k-value was not entered.

• For the unbonded PCC overlay, additional design runs seemed to indicate that the stiffness of the interlayer and the existing PCC was not considered in the calculation of the k-value. It does appear that the base layer was taken into account by the difference in k-values when using a stiff and weak base layer. In addition, the k-values agreed well with the entered dynamic k-value, suggesting that the MEPDG used the entered value for unbonded PCC overlay designs.

• It appeared 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 PCC overlays. It was also apparent that the calculated k-values matched the entered dynamic k-value.

To summarize the preceding discussion, no explicit conclusion could be drawn with respect to defining the layers used in the composite effective dynamic k-values calculated within the MEPDG for each type of overlay design because of the conflicting results obtained. However, it can be concluded that the manually entered k-value were used for unbonded JPCP and bonded JPCP overlay designs but not for the HMA overlay design. No appreciable difference in terms of the design thickness was found among the three design alternatives indicating the reliability of using the backcalculated dynamic (or static) elastic modulus for the PCC layer and the dynamic k‑value for the supporting layers in the MEPDG design. Furthermore, it was also found that the backcalculated k-value that represented the composite stiffness of all layers beneath the slab could be directly entered into the MEPDG without significantly influencing the design thickness for the pavement structure analyzed. However, this does not definitively mean that the MEPDG took the stiffness of the base layer into account in the k-value. It could be either due to the insensitivity of the design thickness on the input k-value or because the unstabilized granular contributes very little to the composite stiffness of all layers beneath the slab.

Case Study 4: Rigid Pavement on Stabilized Base

The most recent FWD test data for section 05-0218 were from 2004. Using the LTPP database, input parameters required for the overlay design were compiled. Not all required data were available for section 05-0218; therefore, sometimes data from other sections of project 05-0200 were used.

Using the FWD test data, the PCC elastic modulus and the dynamic effective k-value were backcalculated for the section. These values were calculated to be 42,580,200 kPa (6.18 million lb/inch²) and 100 kPa/mm (371 lb/inch²/inch), 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 16,536,000 kPa (2.4 million lb/inch²), which was closer to the measured value of 24,804,000 kPa (3.6 million lb/inch²).

As with case study 3, the following three different overlay rehabilitation designs were developed for this project: an HMA overlay, an unbonded JPCP overlay, and a bonded JPCP overlay. The HMA overlay analysis produced an unreasonable overlay thickness of 381 mm (15 inches) for all underlying property scenarios. Several variations in mix properties were analyzed (such as varying binder grades, binder content, air void content, and so on) in efforts to minimize the resultant thickness, but rutting continued to control the design results. If rutting was not considered, an HMA overlay thickness of 102 mm (4 inches) was obtained that successfully met the other performance criteria.

The unbonded PCC overlay analysis produced a 178-mm (7-inch) PCC overlay using approaches based on the following: 1) laboratory-based inputs and 2) backcalculated PCC moduli and k‑values. However, when using a modified PCC elastic modulus and base modulus (accounting for the stabilized base), a much thicker unbonded overlay (381 mm (15 inches)) was obtained. The modified elastic modulus value was as low as 16,536,000 kPa (2.4 million lb/inch²), which did not provide enough structural capacity to adequately carry the future traffic loadings. The bonded PCC overlay analysis produced a 76-mm (3-inch) PCC overlay for all underlying property scenarios. Other observations made during the design analysis process included the following:

• For the HMA overlay design, it seemed impossible to draw a definite conclusion about the constituents contributing to the k-value being reported in the design output (and assumed to be used in the design calculations). The entered dynamic k-value did not appear to be used in the determination of k-value. The difference in the calculated k‑values for low and high base stiffnesses was so slight that it appears that the base layer stiffness was not considered in the calculated k-value.

• With the unbonded JPCP overlay design, a noticeable difference was found in the calculated k-value between 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.

• The modulus of the base appeared to be considered in the calculation of the k-value for bonded PCC overlay designs, which agreed with the assumptions listed in the MEPDG. It was also apparent that the calculated k-values matched the entered dynamic k-values.

• For this case study, a low PCC modulus only influenced the unbonded PCC overlay thickness requirement, more than doubling the required thickness. The PCC modulus did not appear to influence the other overlay design options.

Case Study 5: Rigid Pavement on HMA Pavement

Rigid pavement on HMA pavement design was carried out essentially as a new rigid pavement design. The analysis produced a 178-mm (7-inch) thick concrete overlay for this section based on a 20-year design period.

Based on the results of the sensitivities conducted, it appeared that the MEPDG changed the original flexible pavement structure into an equivalent structure. The equivalent structure consisted of PCC with the same properties as the PCC overlay on top of a base layer with the same properties as the existing HMA; all of this was then supported by Winkler springs with a stiffness equal to the composite k-value established using all layers beneath the existing HMA.

Case Study 6: Composite Pavement

For this project, an HMA overlay thickness of 318 mm (12.5 inches) was required to achieve a 90‑percent reliability, with top-down cracking acting as the controlling criterion. This was unreasonably thick and could be reduced by assuming a higher level of allowable top-down cracking distress or perhaps additional manipulation of the HMA mixture properties.

In this case study, the selection of layer property inputs from backcalculation values had minimal influence on the overall design results. It appeared that the input dynamic k-value was not considered in the design. However, the MEPDG documentation indicates the HMA overlay performance is based on flexible pavement design, so this was consistent in that the design was controlled by the HMA overlay properties. Although the design was controlled by HMA overlay properties, the research team makes the following recommendations:

• The design was insensitive to the trial PCC moduli. Therefore, the use of the dynamic backcalculation adjustment factor (0.8 for the PCC modulus) can continue until new ones are developed or an agency develops specific values.⁽⁶⁾ The MEPDG program appeared to use static PCC elastic modulus values as entered inputs but the output files suggested it reverted to a dynamic value. Additional adjustment of the PCC modulus based on the overall pavement condition could be made, but it does not appear that this influence had an effect unless the pavement was in very poor condition.

• The established modular ratios should be used unless specific testing data are available to determine project specific ratios.⁽²⁷⁾

• The subgrade modulus input should be correlated to the static backcalculated k-value. Both the flexible and rigid design analyses for a composite pavement appeared to use the input subgrade modulus, so this value should be based on the determined support conditions. The program did not appear to use the input dynamic k-value.

• Including an aggregate base layer and determining a corresponding dynamic subgrade k‑value did not appear to have a significant effect on the design results. The addition of a layer would also seem to suggest a change in the climatic adjustment, but it did not appear to be significant, particularly when considering the overall design results.

SUMMARY

Six case studies, using data from in-service pavements, were used to evaluate the way that FWD deflection data were used in the rehabilitation portion of the MEPDG. Specifically, deflection data and backcalculation results were used to characterize the existing HMA, PCC, stabilized and unstabilized bases, and aggregate and subgrade properties in the MEPDG design program. When laboratory testing results were compared with FWD results, the final designs were relatively insensitive to the differences in characterization of existing layer inputs; new material properties tended to control design results. Details of these case studies are found in volume II of this report, but this chapter has presented some of the primary observations and trends. While many of the designs were controlled by new pavement material properties, recommendations on how to use backcalculation data in the MEPDG were discussed.