How Did It Work? Pennsylvania SPS-6 Performance ...

How Did It Work? Pennsylvania SPS-6 Performance at Ten Years—Strategy Evaluation of Concrete Pavement Rehabilitation Dennis A. Morian, P.E. Quality Engineering Solutions, Inc. 7759 Andrews Lane Conneaut Lake, PA 16316 Phone/Fax: (814) 382-3110 [email protected] Laura Coleman Quality Engineering Solutions, Inc. 7759 Andrews Lane Conneaut Lake, PA 16316 Doug J. Frith, P.E. Quality Engineering Solutions, Inc. 888 W. 2nd St., Suite 108 Reno, NV 89503 Phone: (775) 337-2655 Fax: (775) 337-2855 [email protected] Shelley M. Stoffels, D. E., P.E. Dept. of Civil and Environmental Engineering The Pennsylvania State University 212 Sackett Building University Park, PA 16802 Phone: (814) 865-3183 Fax: (814) 863-7304 [email protected] Dan Dawood Bureau of Maintenance and Operations Keystone Building 400 North St. 6th Floor Harrisburg, PA 17120 Phone: (717) 787-4246

Submitted July 31, 2002 in response to Call for Papers for A2B04, TRB Committee on Pavement Rehabilitation Word Count: 7479

TRB 2003 Annual Meeting CD-ROM

Paper revised from original submittal.

Morian, Coleman, Frith, Stoffels and Dawood

2

How Did It Work? Pennsylvania SPS-6 Performance at Ten Years—Strategy Evaluation of Concrete Pavement Rehabilitation ABSTRACT With a long history of constructing jointed concrete pavements, Pennsylvania pursued the construction of an SPS-6 experiment for the “Rehabilitation of Jointed Concrete Pavements”. After ten years of service under very high rural interstate traffic loading, a number of sections require renewal. This paper examines the performance of the eight standard SHRP sections, control section, and three state supplemental sections. The experiment includes a broad range of treatments, including minimal and maximum concrete pavement restoration, thin and thick overlays, crack/break and seat, and rubblization as pretreatment for pavement overlays. Ten years after construction, many of the treatments are in need of major pavement rehabilitation. This provides an opportune time to evaluate the performance of these test sections, both structurally and functionally, to date. Additionally, an evaluation of cost effectiveness clearly shows the importance of performing thorough evaluations of projects for rehabilitation, and the potential benefit of the state supplemental sections employing rubblization of badly damaged jointed concrete pavements as a pre-overlay strategy.




3

How Did It Work? Pennsylvania SPS-6 Performance at Ten Years—Strategy Evaluation of Concrete Pavement Rehabilitation BACKGROUND Until recently, Pennsylvania typically constructed new pavements with portland cement concrete. Since the 1940’s, these pavements have primarily been jointed concrete sections. When the Strategic Highway Research Program (SHRP) began operation in 1987, many of these highways were in need of, or had already experienced the first cycle of rehabilitation. The traditional practice of placing asphalt overlays had been supplemented in the early and mid1980’s with concrete pavement restoration (CPR). Additionally, a wide range of surface preparation strategies prior to overlay were in practice, with an associated range of cost. In 1988, SHRP assembled a group of state experts to design the experiment for the rehabilitation of jointed concrete pavements, SPS-6. Nearly forty states participated in this effort, developing the base SPS-6 experiment design sections. A discussion of practices resulted in selecting a range of pre-overlay preparation with thin (4 in) and thick (8 in) asphalt overlays, as well as minimal and intensive CPR. An untreated control section was included in all the SHRP experiments. A project site for installation of the SPS-6 test section was selected on Interstate 80 in Centre County. The test location selected is in the westbound travel lane in the vicinity of Milepost 155. This location is such that the successive sections progress up a continuous grade. 1988 was also a time when Pennsylvania experience severe pavement rutting statewide in both new and existing asphalt pavements, particularly in overlays of concrete pavement. A SHRP Regional weigh-in-motion site was also installed on I-80 in the relative proximity of the project. This site was designed to collect full traffic data in the travel lane for 300 days or more per year. INTRODUCTION The SPS-6 experiment addresses the performance of various rehabilitation strategies for portland cement concrete pavements. The Pennsylvania SPS-6 project is located in central Pennsylvania on Interstate-80. It lies in the wet-freeze environment area, and falls within the fine-grained subgrade soil cell of the Long Term Pavement Performance (LTPP) experiment (1). In addition to the eight Strategic Highway Research Program (SHRP) test sections, this project includes three supplemental sections. The primary experiment includes minimum, intensive, and break and seat surface preparation treatments. Selected sections then received an asphalt concrete overlay of 4 in (102mm) or 8 in (203mm). The three supplemental sections included a variation in overlay thickness and base preparation, saw slabs at third points plus crack/break and seat with 8 in (203mm) overlay, and 9 ½ in (241mm) and 13 in (330mm) asphalt concrete overlay on rubblized concrete. Reported herein is the performance of these experimental sections after 10 years of service, based upon data provided by the LTPP program (2). PROJECT LAYOUT The PennDOT SPS-6 project is located in the westbound travel lane of the four-lane divided roadway of I-80 near Snowshoe, Pennsylvania, between exits 22 and 23, as illustrated in figure 1. The terrain consists of rolling hills. Consequently, the test sections lie on an upward grade of approximately 4%. The alignment is relatively straight, with areas of both cut and fill within the




4

project. One major bridge structure lies within the construction limits of the project. The traffic at the time of construction was 12,694 AADT with 34% trucks. The westbound roadway consists of two 12 ft (3.65m) wide jointed portland cement concrete lanes with 10 ft (3m) asphalt concrete paved shoulders on the outside, and 4 ft (1.2m) asphalt concrete shoulders on the median side. The existing pavement is 10 in (254mm) of portland cement concrete pavement with 61.5 ft (18.7m) joint spacing on 9 ½ in (241 mm) of crushed limestone subbase on a silt subgrade. The pavement exhibited extensive distress at the joints, including joint seal damage, joint spalling and faulting. The test sections were laid out to minimize traffic disruption on this heavily traveled eastwest highway. Test Sections 420601-420603, 420605, 420606 and the GPS sections, were intended to be those likely go out of service first, permitting rehabilitation without physically disturbing the other experimental sections. In general, paving thickness increased as one progressed through the sections to facilitate paving operations. Routine maintenance consists of joint and crack sealing and patching limited to that normally performed by state maintenance personnel to maintain the safety of the roadway. The experiment sections are described as follows: 601 602 603 604 605 606 607 608 660 661 662

Control with routine maintenance only Minimal surface preparation, no overlay Minimal surface preparation with 4 in (102mm) overlay Minimal surface preparation with saw and seal plus 4 in (102mm) overlay Intensive surface preparation, no overlay Intensive surface preparation with 4 in (102mm) overlay Crack/break and seat with 4 in (102mm) overlay Crack/break and seat with 8 in (241mm) overlay Supplemental: Rubbilize with 9.5 in (241mm) overlay Supplemental: Rubbilize with 13 in (330mm) overlay Supplemental: Saw slabs at third points plus crack/break and seat and 8 in (203mm) overlay

The design schematic for the project is shown in figure 2. The specific treatments and quantities applied to each experimental section are identified in table 1. The crack/break and seat specification requires a crack pattern of 3-6’. The rubblization specification requires all broken pieces to pass 12” maximum size.

Pennsylvania Supplemental Sections Even though Pennsylvania participated in the design of the SPS-6 experiment, considerable work has gone into developing additional rehabilitation treatments suitable for the condition of the I80 pavement at that time (3). The first and second were rubbilization of the existing pavement, with asphalt overlays of 9.5 in (241mm) and 13 in (330mm) structurally designed for ten and twenty years, respectively. The third section was crack and seat of the jointed pavement, with sawcuts sufficiently deep to severe reinforcement at the third points of the slab length. This was overlaid with an 8 in (241mm) structurally designed asphalt overlay.




5

Asphalt Materials The concrete pavement surfaces were overlaid with standard Pennsylvania ID-2 wearing and binder courses using 75-blow Marshall (Heavy Duty) mix designs and 75% crushed aggregate. This was a pre-Superpave rut resisting measure. The in-place air voids of the asphalt sections ranged from 4.3 percent for the binder course to 4.6 percent for the surface course. Asphalt contents were 6.1 percent for the surface course, and 5.7 percent for the binder course. All Marshall stability values exceeded 3000 psi, except one test location in Section 608, which had a value of 2,853 psi. Some thickness variation was encountered from the design target values, as in table 2. Post-construction laboratory testing information of the in-place asphalt mixes was not available. The subbase material consists of crushed limestone, while the subgrade material is silt/clay. Traffic Since the test sections are placed consecutively in the same lane, the traffic is constant throughout all sections. Therefore, a comparison of performance vs. traffic loading can be assumed to be inherent in all the evaluations described. Interstate 80 across Pennsylvania, known as the Keystone Shortway, carries very heavy east-west traffic into New York City. In the twenty years prior to the construction of the SPS-6 Experiment Sections, this section of I-80 carried an annual average of 20,310 ADT, and an estimated average 1,491 KESALs per year. The traffic volume subsequent to the construction has increased from 17,437 ADT in 1993 to a maximum of 24,200 ADT in 1996. However, the volume of trucks has continued to grow from 5,899 in 1993 to 8,776 in 1999, with the percent trucks in the study lane growing from 34% in 1993 to 45% in 1999. This translates to an estimated 14,910 cumulative KESALs per year over a ten-year period. Thirty five percent of the traffic in the test (outside) lane consists of trucks. Traffic prior to rehabilitation grew from 12,899 ADT in 1968 to 17,317 by 1989. The estimated historical traffic for this section of roadway shows that the study lane had already carried 23,647 KESALs between 1968 and 1989. The percentage of trucks varied during those same years from 34-27% as total volumes fluctuated and truck volumes generally grew. Construction The project was constructed during September and October of 1992. Concrete pavement repair work preceded the placement of asphalt overlays. Preliminary testing and material sampling was carried out from August 8-11, 1992. Actual paving took place in late September and early October. Performance Evaluation The performance monitoring data collected under the LTPP Regional Contracts was used to assess pavement performance of the various test sections. Distress One of the most effective means of evaluating structural effectiveness of the various pavement test sections is the assessment of pavement distress development. To expedite the evaluation of distress for the SPS-6 pavement sections, equations developed by the Virginia DOT were applied to the data (4). These equations are based upon distress data and result in various index values. For flexible surfaces, a load related distress index (LDR) and non-load related distress index NDR are computed. For jointed concrete pavements a slab distress rating (SDR) is calculated.




6

Using these equations allows a compilation of the individual distress values into a single index value, which can then be used for comparison. The resulting non-structural (NDR) performance plot (figure 3a) shows that all the overlay sections are expected to reached a terminal serviceability level by 2002, except the two sections constructed by rubbilization with asphalt overlay (sections 660 and 661). The NDR value is primarily based upon the quantity and severity of longitudinal and transverse cracking. The functional performance of sections 660 and 661 remains excellent at this point in time. These comments are based on PennDOT policy of using a Terminal Serviceability Index (TSI) equal to three (based on a 5 point Pavement Serviceability Index (PSI) scale, which corresponds to an NDR value of 60) to trigger action on Interstate highways. Section 660 was designed for a ten-year life, and 661 for twenty years, initially. The distress data indicates no appreciable improvement in the condition of the three original concrete sections (601, 602, 605). All three sections were near the selected terminal serviceability at the time of construction, (having a SDR between 40 and 50) and have continued to deteriorate to a SDR value of 25-30 at the present time as illustrated in figure 3b. From the perspective of structural load related distress (LDR), primarily evaluated on the basis of fatigue cracking, patching, rutting and potholes, all the test sections again reach a terminal serviceability index by 2002 equivalent to an LDR value of 60, except 608, 660, 661 and 662. This clearly shows that rubblization, or break and seat together with a significantly designed overlay thickness produces the best structural performance. Load related distress for the asphalt surface sections is provided in figure 3c. Rutting The development of pavement rutting became one of the most noticeable and critical distresses in the performance of the pavement sections. Sections 601,602, and 605, which retained a concrete pavement surface, exhibit the least rutting, as expected. However, these sections typically developed 0.2-in to 0.3-in ruts over the past ten years, as a consequence of studded tire wear. In comparison, the asphalt-surfaced sections developed rutting ranging from over 0.4 in to 0.75 in. The left (inside) wheel path developed the maximum rut depth in all the overlaid sections. Other than the left wheel path for section 606 (intensive CPR w/ 4-in AC), rutting of the remaining sections was relatively uniform, varying from 0.4 in to 0.6 in. Of these, section 606 developed the most rutting, while section 607 developed the least. Maximum wheel path rut measurements during the past ten years are provided in figure 4. Post-construction testing of the asphalt materials is not available, prohibiting assessment of the mix characteristics. Distress types related to structural rutting are not identified in the data. Consequently, it is likely that the measured rutting distress is the result of either inadequate compaction, or shear flow rutting of the mix. Without the materials testing information, or field investigation, a determination of the cause cannot be made. Deflection Deflection data has been collected for all the test sections under the LTPP program, using Dynatest equipment, in accordance with the LTPP protocol for deflection data collection. (9) This data is compiled in the LTPP database, and has been accessed for analysis. Deflection plots are shown in figure 5a and figure 5b. An analysis of deflection magnitude provides insight into the performance of the test sections. Deflection plots for surface and subgrade layers are provided in figures 5a and 5b, respectively. The concrete-surfaced control and rehabilitation sections, 601,602, and 605 appear




7

to have generally returned to the pretreatment level of deflection magnitude between treatment in1992 and 2001. These plots also show seasonal variation in deflection magnitude resulting from a change in subgrade stiffness. The seasonal increase in deflection magnitude seen during the hot summer months is far greater for the minimal CPR section (602) than for either the intensive CPR or untreated control section. This implies that leaving a pavement undisturbed, or undertaking full repair results in better performance than doing minimal repair work. For the asphalt overlay sections, a significant improvement in deflection before and after treatment can be identified. For most of the sections, regardless of overlay thickness and pretreatment, this improvement is approximately 20 microns. For section 603 it was approximately 35 microns, for 607 approximately 8microns, and for 608 approximately 16 microns. This change is somewhat influenced by the magnitude of response prior to treatment. Greater subgrade support variation provides better opportunity for pavement improvement. All the sections improved to around 20 microns, except 606, which began at a much greater level than most of the others. Deflection magnitude for section 603 has essentially remained constant, or reduced slightly since treatment. For 604 deflections appear to have reduced slightly. Section 606 also shows similar results, a slight increase over time with seasonal variation. Section 607 deflection values show seasonal variation, particularly in the subgrade (sensor7) response. Section 608 shows slight improvement over time. Sections 660 and 662 show a 20-micron improvement in surface deflection, which has remained relatively constant over time. Section 661 shows greater improvement following treatment, and also remains quite good thereafter. These supplemental sections also indicate significant improvement in the subgrade support response over time. The crack/break and seat, and rubblization sections all show significant improvement in the subgrade support condition during the first five years following treatment. Section 608 has continued to improve, while the others remain constant. Roughness The average IRI values for the various sections are shown in figure 6. This information is the result of averaging IRI values in left and right wheel paths for the five multiple runs collected by LTPP. (9) The figure clearly shows the improvement in ride quality that resulted from the rehabilitation work. As expected, the roughness of the control section continued to increase from the initial value of 107 inches/mile to the last reported value of 169 inches/mile. The slight irregularities in the data are attributed to the effects of seasonal climatic variations. The minimal CPR section, 602, benefited only slightly from an initial decrease in roughness (10 inches/mile), and has subsequently increased steadily, though not rapidly to 162 inches/mile by 2001. By 1994, roughness had essentially returned to the pre-treatment level. The intensive CPR section, 605, saw a large initial improvement in roughness. As the pretreatment roughness was among the highest of the test sections, the CPR including diamond grinding reduced roughness by 135 inches/mile. The increase in roughness during the past ten years has been similar to that of the other concrete sections. These three sections currently have the greatest roughness of all the test sections, exceeding 160 inches/mile. The three sections with 4-in asphalt overlay have exhibited nearly identical roughness since rehabilitation. While the pre-treatment roughness of all three varied significantly, from




8

125 to 176 in/mile, all three sections were significantly improved by the treatments to roughness in the range from 65 to 71 in/mile. Section 606 (intensive CPR and 4-in overlay) resulted in an initial roughness reduction of 133 in/mile, second greatest of all sections. Roughness in all three sections has gradually increased since treatment to between 105 and 111 in/mile, providing similar serviceability. Section 604, which incorporated the saw and seal of overlay technique, while initially measured as slightly rougher than the other two 4-in overlays, had the least roughness by 2001. The two crack and seat sections (607 and 608) have provided the best performance of all the original experiment sections in terms of roughness. These sections resulted in the lowest initial roughness after treatment, and continue to have the lowest roughness of the formal experiment sections. The IRI of section 608 (8-in overlay of crack and seat) has remained virtually unchanged (6 in/mile) over the past ten years. The performance of section 607 (crack and seat with 4-in overlay) is nearly as good, showing only a slight increase of about 2 in/mile/year on average. The three supplemental sections have resulted in the lowest IRI over the ten years of performance. Section 662 with an 8-in overlay on crack and seated concrete with mesh sawn at third points of the slab length, has provided performance similar to, but smoother than, that of section 607 (crack and seat with 4-in overlay). Section 660, a 9.5-in overlay on rubblized concrete, performed similarly to the crack and seat with 8-in overlay. Section 661 (rubbilize with 13-in overlay) has performed the best of all the test sections providing a reduction after treatment of 53 in/mile, and increasing less than 6 in/mile over the ten years of performance measured. The IRI in 2001 was still only 69 in/mile. Clearly, the magnitude of initial roughness influences the amount of post-treatment improvement resulting from individual treatment applications. Structural Evaluation In developing the design of SPS experiments, no effort was made to evaluate the structural capacity to support some design traffic for individual projects. Consequently, as a means of comparative analysis, average ESALs per lane were developed from historical traffic data, and used to determine the theoretical (AASHTO-based) ESAL capacity of each section. From this, the AASHTO reliability (5) for the various pavement sections was identified over a ten-year period, based on the measured LTPP lane ESALs (not original design assumptions). All the 4-in asphalt overlays provide reliability of 90%, except for the crack and seat preparation (section 607), which provided less than 50%. Similarly, the crack and seat preparation with 8-in asphalt overlay (sections 608 and 662) provided 85% reliability. The two remaining sections, 660 and 661, were designed for ten- and twenty-year lives originally, using projected traffic. Section 660 (rubblize with 9.5-in overlay) provides 97% reliability, and section 661 provides 99% reliability for a twenty-year design period. These high reliability levels are consistent with the design assumptions used. The rank of reliability values is consistent with the performance observations obtained from the data. The evaluation of the existing concrete pavement indicated that the actual load repetitions experienced prior to rehabilitation exceeded a reasonable design expectation approximately three times (6). The associated fatigue consumption would imply that little continued performance should be expected of the existing slabs, even where patching and load transfer retrofit work was accomplished. Accordingly, the condition of the control section (601) is presently poor.




9

It is noted that all the asphalt pavement overlay sections constructed exceeded the minimum theoretical design thickness. This is a consequence of specification penalties on deficient pavement layer thickness. Analysis of Test Section Cost-Effectiveness In evaluating relative cost-effectiveness of the various treatments, average costs for the various activities on significant projects have been developed for the various cross sections. These costs include asphalt shoulders for all the treated sections. This approach is used since the cost of experimental projects such as the SPS-6 cannot be considered representative of normal production costs. The costs of the various section treatments, including preparation of the existing pavement are provided in table 1. From the information in the tables it is clear that the supplemental sections utilizing break and seat and rubblization strategies have provided the best overall performance, ranking as the top three most cost effective sections. The rubblized section with 13” overlay has provided the best performance over the evaluation period. While several of the other alternative strategies can be expected to have greater costs than these three, they have not provided better performance. These are followed in rank order by section 608, break and seat with 8” overlay. Next in ranking is section 604, minimal CPR with 4”overlay and saw and sealing over transverse joints. This is followed by the 4” overlay sections on minimal CPR (603), The only difference in the ranking provided by the two analysis methods is the order reversal of sections 603 and 607. The CPR sections appear to provide the least cost effectiveness, in this case. The level of initial structural capacity remaining in concrete pavement slabs is significant when contemplating the effectiveness of performing CPR. For projects without adequate remaining life, these treatments may not be cost effective. The previously expended fatigue consumption of the existing pavement slabs has obviously influenced the effectiveness of concrete pavement restoration treatments, since when appropriately applied good performance has been achieved (7). Cost-effectiveness analyses were conducted on the basis of both structural distress performance, and functional performance (roughness). Results were very similar, and the structural evaluation is provided in tables 3 and 3a. Table 3 considers the unit cost per and mile of each treatment, and the decrease in structural performance over the analysis period. The product of the unit cost and decrease in structural index is used to rank the sections. Table 3a provides a separate assessment based upon the equivalent uniform annual cost of the sections, again considering the decrease in structural index, unit cost, interest rate of 3%. Again, the section are ranked in order of cost effectiveness. Based on the results of the cost effectiveness analysis, it is clear that investing in thicker rubblize and overlay is cost effective for heavily traveled highways with aged jointed pavement. These sections result in the best cost effectiveness. It is also clear that the selection of a treatment strategy is greatly affected by the condition of the pavement prior to treatment. The pretreatment condition of the pavement affects this assessment far more for the CPR and thin overlay sections, than for the thicker overlay sections. Previous conclusions regarding the extent of breaking




10

jointed rigid pavements have been reinforced by the performance of these sections. When accompanied by an adequate overlay design, the rubblization of the existing pavement outperforms overlays with extensive slab repair or saw and seal of the overlay, which in turn out performs break and seat. It is clear that the effectiveness of the pavement pre-overlay preparation outweighs the protection afforded by the overlay thickness. CONCLUSIONS A number of conclusions can be drawn from the results of this experimental section review. These are discussed below from both functional and structural performance perspectives. The supplemental sections designed using an AASHTO based procedure, modified for Pennsylvania, have provided the best performance, and remain in service. Those sections constructed in accordance with the original SHRP experiment design using specified overlay thickness provided a lower level of serviceability than the supplemental sections. Typically, when user delay costs are considered, the benefit of pavement sections with longer design life provide even greater cost effectiveness (8). Functional Performance Functional performance is defined as the level of acceptability provided to the traveling public. It addresses the safety and comfort aspects of pavement performance. Pavement roughness provides a reasonable measure of functional performance, along with the NDR distress evaluation. Results from this data provide the following conclusions. • • •

• •

The designed rubblization sections with asphalt overlay have provided the best performance. The crack and seat sections provide the best functional performance of the standard SPS6 test sections. These are followed by the performance of the 4-in minimal overlay sections with similar functional performance as measured in terms of roughness. From this, it can be concluded that the practice of placing a sound asphalt overlay on a jointed rigid pavement can provide a short-term improvement in pavement ride quality, regardless of the extent of preparation. At a more detailed level, the practice of saw and seal of the overlay has resulted in the least roughness after ten years, even though initial roughness was slightly greater than for the two similar sections. The three supplemental sections have provided the best ride quality of all the test sections. Section 661 and 660, rubblize with 13-in and 9-in overlays, have performed best both structurally and functionally of all sections.

Structural Performance Structural performance is defined as the ability of the pavement to sustain repeated loading. It is reflected by the LDR distress evaluation, and deflection response of the sections. Examination of this information has led to the following conclusions. •

Pavement sections developed by a structural design method (supplemental sections) have performed better than the standard experiment sections.



Morian, Coleman, Frith, Stoffels and Dawood • • • •

11

The thin overlay sections have reached a terminal serviceability condition in ten years or less. The effectiveness of the concrete pavement preparation outweighs the effect of the asphalt overlay thickness. Deflection measurements indicate that break and seat, and rubblization improved pavement support conditions for this damaged jointed concrete pavement. The cost effectiveness of the ten and twenty year overlays is significantly better than the standard experiment sections, as should be expected.

RECOMMENDATIONS Based on the conclusions presented above, the following recommendations can be made. 1. For jointed rigid pavements with similar pretreatment condition (showing significant loss of remaining life), rubblization and asphalt overlay provides the best performance of all test sections. 2. The next most cost effective concrete pavement preparation by rank is break and seat with reinforcement severed at slab third points. 3. Although the structural contribution of more extensive pavement breaking reduces as the size of concrete pieces decreases, performance improves as particle size reduces, so long as appropriate overlay thickness is provided. 4. The PennDOT policy of evaluating concrete pavement rehabilitation on the basis of preparation and corresponding overlay design cost has been verified. Pavement rehabilitation strategies should not be predicated on the basis of performance at other locations, or by practice or policy. Site-specific design of pavements is crucial to achieving satisfactory performance, and cost effective investment of highway funds. 5. The performance of CPR can be severely limited by lack of adequate remaining structural life evaluation, which should be performed to assess all candidate projects. 6. Satisfactory functional performance, but limited structural performance, can be achieved for the short term by minimal preparation with thin (4 in) overlays. ACKNOWLEDGEMENTS The authors wish to acknowledge the support provided by the Pennsylvania Department of Transportation, the FHWA Long Term Pavement Performance Division, and the Strategic Highway Research Program in developing the project, and compiling the information analyzed. Data for the state supplemental sections was provided by the North Atlantic Regional Contractor, Stantec. Numerous other individuals, including the construction contractor, H.R.I.Inc., of Snowshoe, PA played a role in the delivery of this valuable information. REFERENCES 1.

Morian, D.A., “Construction Report on SPS-6 Project,” Pennsylvania Department of Transportation, Prepared by North Atlantic Regional Contractor, Pavement Management Systems Limited for Federal Highway Administration, Long Term Pavement Performance Division, May 1995.

2.

SPS-6 LTPP Data, North Atlantic Regional Contractor, Stantec (CD-ROM)




12

3.

Morian, D. A.,"Rubblization of Concrete Pavements in Pennsylvania”. The Pennsylvania State University, The Graduate School, Department of Civil and Environmental Engineering, December 1994.

4.

Sami, N., McGhee, K.H., ”Condition of the Pavement – 1998, Interstate and Primary Highways”, Virginia DOT Pavement Management Program

5.

“AASHTO Guide for Design of Pavement Structures, 1993”. American Association of State Highway and Transportation Officials, 444 N. Capital Street, N.W., Suite 249, Washington, D.C. 20001.

6.

“Pavement Policy Manual”, Pennsylvania Department of Transportation, Publication 242, 1989.

7.

Morian, D.A.and Cumberledge, G.C. “Techniques for Selecting Pavement Rehabilitation Strategies, Pennsylvania Case Studies,” Transportation Research Board Record No. 1568, Pavement Design Management, and Performance, Pavement Rehabilitation and Design,” National Academy Press, Washington, D.C. 1997.

8.

Morian, D.A., Gibson, S., “Maintenance, Rehabilitation, and Reconstruction of High Volume Rigid Pavements, Draft”, for NCHRP Project 10-50, 1998.

9.

LTPP web address : http://www.tfhrc.gov/pavement/ltpp/ltpp.htm




13

LIST OF TABLES Table 1 Table 2 Table 3 Table 3a

Pennsylvania SPS-6 Matrix Asphalt-Surfaced Section Layer Thicknesses Rank of Cost-Effectiveness Ratios EUAC of Life Extension Approach

LIST OF FIGURES Figure 1: Figure 2: Figure 3a: Figure 3b: Figure 3c: Figure 4: Figure 5a: Figure 5b: Figure 6:

Location Map SPS 6 Design Schematic Non-load Related Distress Rating Slab Distress Rating Load Related Distress Rating Maximum Wheel Path Rutting Surface Deflection Sensor 7 Deflection Average International Roughness Index




14

Table 1 Pennsylvania SPS-6 Matrix Section ID 601 602 603 604 605 606 607 608 660 661 662 Total

Load Full Partial Transfer Depth Depth Treatment Preparation Underdrains Underseal Restoration Patch Patch None None Min. CPR Min. CPR 7 Min. CPR w/4-in AC Min CPR Y 3 slabs Min. CPR w/4-in Saw & Seal AC Min. CPR Y Y 2 2 Max. CPR CPR Y Y 8 8 Max. CPR w/4-in AC CPR Y Y 3 10 C&S w/4-in AC C&C Y C&S w/8-in AC C&S Y Rubblize w/9.5-in AC Rubblize w/13-in AC C&S, Saw 1/3 Points, 8-in AC C&S, Saw 18 27




15

Table 2 Asphalt-Surfaced Section Layer Thicknesses Existing Asphalt Asphalt Asphalt Total Design Section Granular Existing Overlay Overlay Overlay Asphalt Overlay ID Subbase (in) PCC (in) Surface (in) Binder (in) Leveling (in) Overlay (in) Thickness (in) 603 9 10.1 1.7 2.4 3.8 4 604 9 10.2 1.9 2.3 4.2 4 606 9 10.0 1.9 2.8 4.3 4 607 9 10.1 1.8 2.5 4.1 4 608 9 10.1 1.8 5.7 1.4 8.5 8 660 9 10.6 1.7 6.4 0.9 9.6 9.5 661 9.1 9.9 1.7 10.7 1.2 13.3 13 662 9 10.1 1.8 5.8 8.2 8




16

Table 3 Rank by Cost-Effectiveness Ratio Table 3: Cost Effectiveness of Section Performance Section Cost ($*10^4/ln-mile) Delta LDR B* C Rank 661 5.95 7 41.65 1 660 4.94 10 49.4 2 662 4.54 14 63.56 3 608 4.52 19 85.88 4 604 5.34 17 90.78 5 603 5.32 29 154.28 6 607 2.46 65 159.9 7 606 7.49 24 179.76 8 602 3.98 49 195.02 9 605 6.15 64 393.6 10

Table 3a: EUAC of Life Extension Approach Table 3a: EUAC of Life Extension Approach Assume loss of 30 on LDR to be limit of life extension. Assume LDR loss was over 7 years. Assume i=3%/year. Section Cost ($/ln-mile) Delta LDR Life Extension (yrs) EUAC Rank 661 $59,500.00 7 30.0 ($3,036) 660 $49,400.00 10 21.0 ($3,205) 662 $45,400.00 14 15.0 ($3,803) 608 $45,200.00 19 11.1 ($4,865) 604 $53,400.00 17 12.4 ($5,237) 607 $24,600.00 65 3.2 ($8,103) 603 $53,200.00 29 7.2 ($8,283) 606 $74,900.00 24 8.8 ($9,860) 602 $39,800.00 49 4.3 ($10,035) 605 $61,500.00 64 3.3 ($19,960)


1 2 3 4 5 6 7 8 9 10




17



18

Figure 2




19

Fig 3a Non Distress Related Rating 100

90

80

70 Section 603 Section 604

60 NDR

Section 606 Section 607

50


40


30

20

Note: Actual data collection dates, not equally spaced

10

0 01/31/93

06/15/94

10/28/95

03/11/97

07/24/98

12/06/99

04/19/01

09/01/02

Date




20

Fig 3b Slab Distress Rating 80

70

60

SDR

50 Section 601 40


30

20


10

0 12/23/88

5/7/90

9/19/91

1/31/93

6/15/94

10/28/95

3/11/97

7/24/98

12/6/99

4/19/01

Date




21

Fig 3c Load Related Distress Rating 100 90 80 70 Section 603 Section 604

60 LDR


50


40


30 20


10 0 01/31/93

06/15/94

10/28/95

03/11/97

07/24/98

12/06/99

04/19/01

Date




22

FIGURE 4 Maximum Wheel Path Rutting 0.80

0.70

0.60

Rut Depth (in.)

0.50

Left Wheel Path

0.40

Right Wheel Path

0.30

0.20

0.10

0.00 601

602

603

604

605

606

607

608

660

661

662

Section




23

Figure 5a Surface Deflection Date 5/7/90 0

9/19/91

1/31/93

6/15/94

10/28/95

3/11/97

7/24/98

12/6/99

4/19/01

9/1/02

50

100 Section 601

Deflection (microns)

150


200


250

Section 606 Section 607 Section 608

300


350

Section 662

400 Note: Actual data collectin dates, not equally spaced

450

500




24

Figure 5b Sensor 7 Deflection Date 5/7/90 0

9/19/91

1/31/93

6/15/94

10/28/95

3/11/97

7/24/98

12/6/99

4/19/01

9/1/02

50

100 Section 601

Deflection (microns)

150


200


250


300


350

Section 662

400 Note: Actual data collection dates, not equally spaced

450

500




25

Figure 6 Average International Roughness Index 350

300

Section 601

Roughness Index (in/mile)

250

Section 602 Section 603 Section 604 Section 605

200


150


100


50

0 12/23/88

5/7/90

9/19/91 1/31/93 Treatment Application

6/15/94


10/28/95

3/11/97

7/24/98

12/6/99

4/19/01

9/1/02

Date
