Significant Findings from Full-Scale Accelerated Pavement Testing (2004)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

185 APPENDIX G Summary of Accelerated Pavement Testing Objectives and Applications Table G1 Structural Design Applications (refer to chapter two) Acronym/Location Objectives Applications Proof trial ALF. Evaluate macadam base. Structural adequacy of Somersby pavement confirmed. ALF confirmed as reliable and effective. Compare thin (200 mm) and standard (300 mm) cement-treated base (CTB). Failure mode was debonding of CTB layers, ingress of water at interfaces, and subsequent erosion of bottom of upper layer, leading to break-up of top CTB layer. Compare pavements with and without a bitumen heavy-cure coat interlayer and constructed in one lift instead of two or three lifts. Failure mechanism of adjacent National highway found to be the same. Major changes to construction practice implemented. ALF Australia Determine the relative fatigue characteristics of various types of asphalt rehabilitations including geotextiles, modified asphalt, and high asphalt content mixes in the “tensile” zone applied to a distressed pavement. Overlay trials demonstrated improved performance, for a modified binder compared with conventional binder; a significant impact on treatment for urban rehabilitation. SAMI successfully delayed crack propagation. Inaugural testing: four thicknesses of unbound granular pavements under chip seals. Proving trials and assessment of traditional pavements. Lime-stabilized subbases: three thicknesses. An experiment with lime-stabilized subbase course showed that an increase in thickness from 150 to 250 mm would increase life by 15 times (Pidwerbesky 1995a). CAPTIF New Zealand Life-cycle performance of a thin- surfaced unbound granular pavement. Assessment of traditional pavement configuration. FHWA–PTF USA To evaluate accuracy of AASHO designs used. Good agreement was reached on one section; the second section failed prematurely by cracking. HVS South Africa Reduction of pavement cost by improved designs. New designs incorporated in the catalogue resulting in construction cost savings. Comparison of pavement configurations. Effective ranking procedure developed. ISETH Switzerland Evaluation of design procedure. Existing practice shown to be conservative. Performance of doweled concrete pavements and different cements. Revision of design procedure. Trafficking of three thick structures (up to 200 mm) of bitumen-treated materials. Adjustment of the circular test track. Evaluation of four structures: CTM/CTM, CTM/CTM, BTM/CTM, and CTM/UTM. Heavily traveled pavements with road bases treated with hydraulic binders. Comparison of bitumen-treated materials on WRM and cement-treated sand subbases. Demonstration of separating layer of WRM to reduce reflective cracking. LCPC France Two very thin rolled asphalt overlays assessed. For maintenance motorway pavements.

186 Table G1 (Continued) Reduce pavement costs by using stabilized bases and thin asphalt surfacing. A 90–150-mm asphalt layer over a stabilized base was equivalent to 200-mm full-depth asphalt. RIOH–ALF China Compare the performance of 10 different asphalt pavement configurations. New asphalt pavement structures were recommended for the national specifications. TRL United Kingdom Forest road design. Pavement design methods for specialized log traffic developed. Table G2 Vehicle–Pavement–Environment Interaction (refer to chapter three) Acronym/Location Objectives Applications To determine the axle load equivalency for a typical crushed-rock pavement, 300 mm thick, subjected to accelerated loading under ALF single-axle dual- wheel loads of 40, 60, and 80 kN. Load equivalency factors of 10 and 8, respectively, were estimated for a maximum surface deformation of 10 mm. ALF Australia Compare the performance of crushed- rock pavements, constructed at different moisture/compaction conditions at sealing and after drying-back and wetting-up after sealing. At Beerburrum the low-plasticity, highly permeable crushed-rock bases, when placed on a CTSB having a 3% cross-fall, quickly dried back in the dry environment operating at that time. Comparative rutting: dual and wide- base single tires. Tire studies, which compared single low-profile, wide-based, radial ply to standard dual tires, and showed 92% greater rutting for the singles. Strain response of subgrades and unbound granular pavements: wheel load, tire pressure, and tire type. Development of subgrade strain criteria for design. Dynamic wheel loads and pavement wear: single-unit and multi-leaf spring suspensions. Vehicle–pavement interaction studies. CAPTIF New Zealand DIVINE project; air bag and multi-leaf suspensions. Vehicle–pavement interaction studies. Evaluation of frost protection design. Existing practice shown to be conservative. Evaluation of drainage layer. Shown to be very effective in freeze/thaw environments. EPFLa Switzerland Effects of temperature and loading pattern. Dependence of rutting on temperature and transverse distribution of wheel path shown. FHWA–PTF USA To assess the impact of tire pressure on pavement response and performance. Tire pressure effects shown to be less significant than load and temperature on flexible pavement response and performance. Characterization of tire/pavement interface stresses. As part of the HVS program it was possible to develop a 3-D load cell. This provided a means of accurately characterizing and modeling the tire/pavement interface. HVS South Africa Improved understanding of response and performance. Improved load equivalency factors.

187 Table G2 (Continued) Comparison of the effects of 13-t and 10-t axles and of an 80 mm against two successive 40 mm overlays. To design of WRM and CTM road bases and overlay practice. To compare: design methods and performance models; tandem axles versus standardized European axle. Working group 14 of the OECD tested two flexible and one semi-rigid pavement. LCPC France Observation of rutting behavior of bituminous concrete. Evaluation of high-performance mixes, influence of speed-temperature and wheel load effects. Determine structural damage of tri- axles. Approximate load equivalency factor of 2.6 established, varying with load and pavement configuration. Evaluation of free draining base course. Shown to be compaction-sensitive and best used deep in a pavement. PTI USA Structural coefficient of surface mixes. New materials and mixes performed in a similar manner to existing materials. Determine the failure mechanism investigation of axle-load equivalencies for this type of pavement. Surface seal maintenance was important to prevent water entering cracks and precipitating failure. RIOH–ALF China Determine the relative damaging effects of different axle loads in terms of structural failure and surface rutting. The current equivalent “exponent” was increased. TRL United Kingdom Single versus dual tire damage evaluation. Single tires cause twice the damage on the thick pavements typical of the UK and Europe. Table G3 Materials and Tests (refer to chapter four) Acronym/Location Objectives Applications Validate ERDC–GSL–WES tentative classification for lateritic gravel for road and airfield pavements. Performance of both materials constructed to both compaction standards was excellent when tested soon after construction. When the pavements were retested after a wet season, the performance of the good material was still satisfactory but the poor material failed very quickly under traffic. Determine the effects of bitumen and bitumen/cement stabilization on the performance of a reconstructed high- quality crushed-rock pavement. Recycling existing crushed-rock bases by stabilizing with 2% bitumen and 2% cement could improve performance and resistance to water penetration. Performance was very sensitive to additive content and compaction level. Evaluate high-quality dense crushed- rock base pavement for heavy traffic. Structural adequacy of Benalla pavement confirmed. Compare performance of this pavement (double-chip seal surface) with similar pavement tested at Somersby (asphaltic concrete surface). Heavy compaction and prompt seal maintenance shown to be essential. ALF Australia Assess the use of unbound and stabilized slag as base materials. Specification for road base materials adjusted to permit wider use of slag materials.

188 Table G3 (Continued) Compare performance of different thicknesses of two qualities of recycled marginal unbound bases. When subgrade in existing pavement is adequate for the design traffic, existing sandstone base can be reconstructed to support the seal and provide additional load capacity. Reconstructed sandstone base thicker than 125 mm may not enhance pavement performance, particularly for traffic with low axle loads (<40 kN). Recycling existing sandstone bases by stabilizing with 2% bitumen and 2% cement could improve performance and resistance to water penetration. Validate ERDC–GSL–WES tentative classification for lateritic gravel for road and airfield pavements. Performance of both materials constructed to both compaction standards was excellent when tested soon after construction. When the pavements were retested after a wet season, the performance of the good material was still satisfactory, but the poor material failed very quickly under traffic. Quantify relative rut resistance of new and conventional mixes. The performance of the AUSTROADS mix was contrary to expectations because laboratory creep testing had indicated that this mix was more rut- resistant than the control mix. ALF Australia To quantify rut resistance of experimental mixes developed by industry compared with a conventional mix of known performance, both in the field and under accelerated loading. The results associated with the control mix are yet to be analyzed. To compare rutting of a dense-graded asphalt under channelized and normal traffic. Channelized traffic doubled the rutting surface rut rates. This compared well with previous HVS trials at about 100,000 repetitions. CAL–APT California To compare the fatigue performance of a rubber–bitumen mix to a conventional mix. Life to cracking failure compared well with that calculated by the Caltrans overlay design procedure, a reduction of 50% in layer thickness is justified for the rubber–bitumen mix. Effect of particle shape and gradation on base course. Performance: nine pavements. New Zealand specifications currently limit the rounded aggregate content in base course to a maximum of 30%. The research showed that, while the best performance was obtained with this mix content, up to 50% could be tolerated (Pidwerbesky 1995a) and that mixes with greater than 70% rounded aggregate could not be compacted. CAPTIF New Zealand Asphalt mixes with four modified and two conventional binders. Six asphaltic concrete layers compared high and conventional stiffness mixes (Pidwerbesky 1995a). The sections showed little deterioration and the tests were truncated at a surface deformation of 4 mm; at this stage the stiffer asphalt mixes were performing better than a thicker layer with conventional binder. HVS South Africa Use of nonstandard materials. Use of sands and sandstones in stabilized subbase specifications for: emulsion-treated bases, use of blast furnace slag, use of water- bound macadam.

189 Table G3 (Continued) Investigation of porous asphalt. Performance of bitumen–rubber porous asphalt, with void contents in excess of 20%, investigated in terms of deformation. Findings used in porous asphalt design manual (SABITA 1996). HVS South Africa Comparison of HVS predicted behavior with actual pavement performance. Performance of a road 15 years after HVS testing investigated. Good agreement was found with this study, the first such comparison of any extent. Number of important issues relate to effective long-term pavement performance monitoring. JHPC Japan To compare low-volume road base courses. Base course materials were ranked in order of performance; a manual for low-volume roads was prepared. Tests of four free-draining coated materials. Ranking of performance in relation to rutting, cracking, drainage, and frictional characteristics. Evaluation of four structures of untreated materials having very different characteristics. Funded by aggregate producers. Evaluation of special materials. To acceptance of commercial mixes. Study of high-performance mixes and fatigue behavior. Comparison with normal mixes and laboratory tests. Observation of rutting behavior of bituminous concrete. Evaluation of high-performance mixes, influence of speed-temperature, and wheel load effects. Comparison of hot and cold bituminous mixes. Development of maintenance techniques. LCPC France Test commercial emulsion-bound composite material. Validated performance. Determine in situ moduli from surface deflection. Moduli higher than laboratory estimates and affected by many variables. PTI USA Evaluation of free-draining base course. Shown to be compaction-sensitive and best used deep in a pavement. Investigate the performance of low-cost, heavy-duty pavements with lime- stabilized soil bases. Asphalt surfaced lime-stabilized soil base pavement had a life in excess of 6 million ESALs. RIOH–ALF China Determine the fatigue performance of stabilized base materials using tensile strains. Pavement life models using laboratory tensile (beam) tests did not correlate closely to behavior under ALF.

190 TABLE G4 Modeling (refer to chapter five) Acronym/Location Objectives Applications ALF Australia Establish relationships between back- calculated asphalt stiffness and CTCR modulus, determined from FWD deflection bowl data, pavement temperature, and the severity and extent of surface cracking. Asphalt fatigue relationships derived. Potential benefits much higher as use of heavy- duty asphalt pavements in urban applications increases. To compare estimated and measured pavement responses. There was general agreement between peak deflection, back-calculated moduli, and strain data. Cracking was not a good indicator of failure unless the cracks propagate to the surface. FHWA–PTF USA To establish load equivalencies. Not yet complete. Improved understanding of response and performance. Improved mechanistic models. HVS South Africa Improvements in modeling permanent deformation. Contributions of the various rock types in the pavement structure. To compare: design methods and performance models; tandem axles versus standardized European axles. Working group 14 of the OECD tested two flexible and one semi-rigid pavement. LCPC France To test cold mix. Two phases were used. The first load cycles were applied at 45 kph with a reduced load. The second trafficking took place after a few months with full load and speed. This allowed for maturing. Develop structural equivalency factors. Layer coefficients recommended. PTI USA Develop estimates of remaining service life. Remaining service life can be predicted from Surface Curvature Index; effect of layer thickness considered in terms of structural number. RIOH–ALF China Compare field measurements of stresses and strains and evaluate the theoretical model. A theoretical linear elastic model and back- analysis procedure were amended to allow for conditions from full to no layer bonding. Table G5 Rehabilitation, Construction, and Maintenance (refer to chapter six) Acronym/Location Objectives Applications Compare pavements with and without a bitumen heavy-cure coat interlayer and constructed in one lift instead of two or three lifts. Failure mechanism of adjacent National highway found to be the same. Major changes to construction practice implemented. ALF Australia Compare the performance of two thicknesses of gravel pavements and the geotextile-reinforced seal pavements. Guidelines for the design, construction, maintenance, and management of geotextile- reinforced seal pavements prepared. Major long-term benefit is more effective use of local materials and scant resources in a location where it is imperative to maintain road links during the wet season.

191 Table G5 (Continued) Compare performance of different thicknesses of two qualities of recycled marginal unbound bases. When subgrade in existing pavement is adequate for the design traffic, existing sandstone base can be reconstructed to support the seal and provide additional load capacity. Reconstructed sandstone base thicker than 125 mm may not enhance pavement performance, particularly for traffic with low axle loads. Recycling existing sandstone bases by stabilizing with 2% bitumen and 2% cement could improve performance and resistance to water penetration. Compare the performance of crushed- rock pavements, constructed at different moisture/compaction conditions at sealing and after drying-back and wetting-up after sealing. At Beerburrum the low-plasticity, highly permeable crushed-rock bases, when placed on a CTSB having a 3% cross-fall, quickly dried back in the dry environment operating at that time. ALF Australia Establish the performance of deep-lift recycled pavements using a slag/lime binder over subgrades of relatively low and relatively high strengths. At Cooma, deep-lift recycled pavements tested on a low-strength subgrade (CBR = 4) had fatigue lives at least twice that estimated for the adjacent National highway. The findings suggest that this type of pavement recycling is suitable for moderate rural arterial traffic. Development of low-risk maintenance strategies. Better use of cemented materials for heavy traffic roads. Use of crushed-stone bases for heavy traffic— 15–20 ESALs. Reduced cost of bituminous bases for heavy traffic. Extended use of natural gravel. Importance of proper maintenance. Evaluation of rehabilitation. Recycling of asphalt pavement. Advantages of early rehabilitation of crushed- stone bases. Prediction of remaining life and recommended rehabilitation procedure for concrete pavement affected by alkali–aggregate reaction. Rehabilitation of a deformed bitumen premix. Materials test methods. HVS testing resulted either directly or indirectly in the development or refinement of several material test methods and associated design criteria. It included the erosion test for cementitious materials, the crack movement simulator for evaluating crack reflection, and the refinement of the dynamic creep test for asphalt deformation. The link between laboratory test results and APT test results is vital to ensure full benefit from the program. HVS South Africa Comparison of bases constructed by labor-enhanced techniques. Labor-enhanced, or labor-intensive, construction techniques. The construction covered different bases including penetration macadam, emulsion-treated natural gravel, slurry bound macadam, and clinker ash.

192 Table G5 (Continued) HVS South Africa Rehabilitation measures for cemented- base pavements. A long-term HVS investigation into the selection of rehabilitation measures for lightly cemented-base pavements concluded (Steyn et al. 1997). ISETH Switzerland Evaluation of failures. Construction quality shown to be a common factor. Comparison of hot and cold bituminous mixes. Development of maintenance techniques. LCPC France Comparison of very high modulus bituminous and cement-treated materials. Rehabilitation techniques. Rigid pavement repair. Techniques validated. Evaluation of recycled asphalt. Technique validated. PTI USA Overlay design. Simplified mechanistic procedure based on Road Rater deflections. Trench reinstatement techniques. Present methods shown to be conservative and thus legal requirements realistic. Ecopave evaluation. New concrete paving technology proven and patented. TRL United Kingdom Nu-pave evaluation Thin fiber-reinforced concrete overlay technique designed to crack without “failing.” Table G6 Pavement Engineering Applications and Issues (refer to chapter seven) Acronym/Location Objectives Applications Compare the performance of two thicknesses of gravel pavements and the geotextile-reinforced seal pavements. Guidelines for the design, construction, maintenance, and management of geotextile- reinforced seal pavements prepared. Major long-term benefit is more effective use of local materials and scant resources in a location where it is imperative to maintain road links during the wet season. Establish relationships between back- calculated asphalt stiffness and cement- treated crashed-rock (CTCR) modulus, determined from FWD deflection bowl data, pavement temperature, and the severity and extent of surface cracking. Asphalt fatigue relationships derived. Potential benefits much higher as use of heavy- duty asphalt pavements in urban applications increases. ALF Australia Assess the use of unbound and stabilized slag as base materials. Specification for road base materials adjusted to permit wider use of slag materials. FHWA–PTF USA To establish load equivalencies. Not yet complete. Improved understanding of response and performance. Improved mechanistic models. Improvements in modeling permanent deformation. Contributions of the various rock types in the pavement structure. HVS South Africa Enhancement of PMS procedures. Visual cracking was adopted as a trigger for resealing. The HVS showed that cracking in the thin surfacing used in South Africa would lead to rapid pavement deterioration with ingress of water, before any significant change in deflection.

193 Table G6 (Continued) JHPC Japan To rank the durability of overlay treatments Findings were used in a maintenance manual. To test cold mix. Two phases were used. The first load cycles were applied at 45 kph with a reduced load. The second trafficking took place after a few months with full load and speed. This allowed for maturing. Evaluation of four structures of untreated materials having very different characteristics. Funded by aggregate producers. LCPC France Evaluation of special materials. Towards acceptance of commercial mixes. Skid resistance. Calibration service provided. Develop structural equivalency factors. Layer coefficients recommended. Develop estimates of remaining service life. Remaining service life can be predicted from Surface Curvature Index; effect of layer thickness considered in terms of structural number. PTI USA Roughness measurement. Calibration technique developed using sinusoidal blocks. Reduce reflective and thermal cracking of asphalt pavements by using a rubber asphalt interlayer. Life to the onset of cracking was much improved. Evaluate the performance of the prototype desert pavements. A design for a desert highway was recommended and has been used. RIOH–ALF China Compare field measurements of stresses and strains and evaluate the theoretical model. A theoretical linear elastic model and back- analysis procedure were amended to allow for conditions from full to no layer bonding. Sidewalk damage by heavy vehicles. Damage to sidewalks evaluated because of legal liability issues. TRL United Kingdom Industrial by-products. Specifications developed to permit use of several materials shown to be structurally adequate and environmentally safe. (Source: Metcalf 1996, with some referenced supplements.)