Composite Pavement Systems, Volume 2: PCC/PCC Composite Pavements (2013)

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.

21 Introduction The field composite pavement sections used in the structural modeling included a combination of special research sections in the United States and Canada and regularly constructed projects in Europe. 1. Regularly constructed projects in Europe: a. Austria is one of the most experienced countries with regard to construction of two-lift PCC/PCC compos- ite pavements, which is their standard design for PCC pavements. Two projects on A1 in Austria were sur- veyed as part of the R21 European trip. These sections were constructed in the mid-1990s. b. Germany is also very experienced in the construction of composite pavements. As part of the R21 European trip, a section of PCC/PCC composite pavement, con- structed in 1995, was surveyed on the A93. c. The Netherlands also routinely constructs PCC/PCC composite pavements. A section on N279, constructed in 2000, was surveyed as part of the R21 European trip. 2. Specially constructed research sections: a. MnROAD included two PCC/PCC composite sections on I-94 under heavy truck traffic and severe weather conditions: • Cell 71 was constructed in May 2010 under SHRP 2 R21. This section was 3-in. PCC over 6-in. PCC (using RCA). Joints were spaced at 15 ft and 1.25-in. diameter dowels were used at the transverse joints. This section exhibited no distress after a full year of 1 million heavy trucks. • Cell 72 constructed in May 2010 under SHRP 2 R21. This section was 3-in. PCC over 6-in. PCC (using low-cost high fly ash content PCC). Joints were spaced at 15 ft, and 1.25-in. diameter dowels were used at the transverse joints. This section exhibited no distress after a full year of 1 million heavy trucks. b. Experimental PCC/PCC sections have been constructed on State Route 45 in Fort Myers, Florida; US-75 in Rock Rapids, Iowa; K-96 in Haven, Kansas; I-70 in Salina, Kansas; I-75 in Detroit, Michigan; and A15 in Montreal, Quebec, Canada. These sections constructed in the last four decades differ with respect to mix designs for the PCC layers and surface textures. These sections gener- ally were constructed as relatively short experimental sections (except I-70 in Kansas, where the total project length with seven different surface textures is several miles long) with special design features and not as part of routine construction in these states. Figure 2.1 shows the geographic locations of the PCC/PCC composite sections. The sections can be seen to offer a reason- able spread across different geographic and climatic conditions in the United States and Canada. The sections in Germany, Austria, and the Netherlands are also shown. These PCC/PCC composite pavement sections include several different types of surface textures, including EAC, conventional diamond grinding, next-generation diamond grinding, longitudinal tining, longitudinal grooving, Astroturf drag, and Astroturf drag and longitudinal grooving. These sections show a range of designs, including the following: • The thin PCC surface layers range in thickness from 1.5 to 3.5 in. The top layer PCC varied considerably from one section to another with respect to aggregates (types, hard- ness, gradation, and so forth), cement content, and use of SCMs such as fly ash and pozzolan, use of admixtures, and other mix properties. • The thicker PCC lower layers range in thickness from 6 to 11.8 in. The lower layer PCC varied considerably with respect to mix designs and included conventional PCC, PCC with RCA, PCC with high fly ash content, PCC with low cement content (econocrete), and inclusion of ASR susceptible aggregates. C h a p t e r 2 PCC/PCC Test Sections

22 test Sections at MnrOaD Introduction In May 2010, two full-scale PCC/PCC test sections were constructed on I-94 at MnROAD to emulate best practices of constructing PCC/PCC composite pavements. Before the con- struction of the mainline test sections, a 200-ft two-lane test strip was constructed at the MnROAD facilities. A summary of the construction of the test section from initial site grading and aggregate base compaction to PCC placement and instrumen- tation installation is presented in this section. Details of each of these topics are included in Appendix F. Figure 2.2 shows the location of the MnROAD test section relative to Minne- apolis. An aerial view of a portion of the MnROAD facility is shown in Figure 2.3. Design and Specifications The project consisted of recycling an existing concrete pavement; the coarse aggregate (RCA) from the recycled pavement was used to construct the lower PCC layer for one of the PCC/PCC test sections. PCC/PCC sections constructed at MnROAD were designed to feature a 3-in. high-quality PCC layer over a 6-in. low-cost or RCA PCC lower layer. The 1 Netherlands 2 Austria 1 Germany 5 Florida Figure 2.1. Map showing geographic dispersion of PCC/PCC sections. Source: © 2012 Google. Figure 2.2. Location of the MnROAD PCC/PCC on I-94 near Albertville, approximately 40 miles northwest of Minneapolis.

23 term “low-cost” signifies that the PCC design was such that the lowest possible amount of cement and most inexpensive coarse aggregates were used by the contractor. In addition, various textures were considered for the surface PCC layer. Because of its potential with respect to durabil- ity, texture, and noise characteristics, an EAC was chosen. The EAC was constructed for both PCC/PCC test sections. How- ever, the first 475 ft of the EAC was considered the “learning area,” and the EAC texture was later diamond ground using conventional grinding in the passing lane and next-generation grinding in the driving lane. The designs are summarized in Table 2.1 and Figure 2.4. Note: The mainline I-94 traffic is diverted to the center lanes during construction and testing. Low-volume test loop Location of PCC/PCC sections on I-94 mainline Figure 2.3. Aerial view of MnROAD facility and location of PCC/PCC test sections. Table 2.1. PCC/PCC Designs for MnROAD Sections Section PCC over RCA PCC (Cell 71) PCC over Low-cost PCC (Cell 72) Upper PCC Thickness 3 in. 3 in. Mix High-quality portland cement (~616 lb/yd3) plus 109 lb/yd3 (15%) fly ash Class C (FAC) High-quality portland cement (~616 lb/yd3) plus 109 lb/yd3 (15%) FAC Coarse aggregate Crushed granite (maximum size 3⁄8 in.) Crushed granite (maximum size 3⁄8 in.) Lower PCC Thickness 6 in. 6 in. Mix Low-quality portland cement (~360 lb/yd3) plus 240 lb/yd3 (40%) FAC Low-quality portland cement (~240 lb/yd3) plus 360 lb/yd3 (60%) FAC Coarse aggregate 50% RCA, 50% Minnesota DOT Class A, maximum aggregate size 1.25 in. 100% Minnesota DOT Class A, maximum aggregate size 1.25 in. Base 8 in., Class 5 unbound 8 in., Class 5 unbound Subgrade Clay Clay Joint spacing 15 ft 15 ft Doweling 1.25 in. placed on baskets at total PCC middepth 1.25 in. placed on baskets at total PCC middepth Surface texture Conventional diamond grinding, next-generation diamond grinding EAC, conventional diamond grinding, next-generation diamond grinding

24 Construction of the Test Sections The construction project was awarded to C. S. McCrossan of Maple Grove, Minnesota. WSB and Associates, Inc. was responsible for the administration of the construction contract and the inspections. Table 2.2 shows a timeline of the major steps involved in the construction process. Because of the unique nature of this project, special pro- visions were used as part of the bid package to modify the DOT’s existing specifications. The special provisions included 1. Salvage concrete pavement: Specifications for the salvage operation to recycle and reuse coarse aggregate from existing on-site concrete pavement. 2. Structural concrete: Specifications for the concrete mix design and aggregate gradations for both PCC layers and concrete design details. 3. Concrete curing and texturing: Specifications for the cur- ing and texturing of the PCC surface particularly to obtain the EAC surface texture (application of curing/retarding compound, brushing, and so forth). 4. Concrete pavement joints: Specifications covering details of saw cutting the joints in the surface PCC layer. 5. PCC/PCC composite pavement operation: Sequence of paving activities for the construction of PCC/PCC com- posite pavements. Recycling Operations The recycling operations consisted of breaking, removing, transporting, crushing, washing, screening, and stockpiling the concrete pavement material from an existing MnROAD cell to be used as coarse aggregate in the recycled concrete mix. The concrete portions of the existing cells were broken with a guillotine crusher (Figure 2.5), removed (Figure 2.6), and transported to a crushing location. The crushing method and system determines some of the qualities of the RCA, such as mortar content and the gradation. An increase in the number of crushing processes reduces the mortar content (Sanchez de Juan and Gutierrez 2009). As specified, all joint material, reinforcing members, and other inert materials (such as wood) were separated from Table 2.2. Construction Timeline for Major Tasks Major Task Date Salvage and recycling operations April 12–16, 2010 Trimming and grading of subgrade April 19–22, 2010 Aggregate base placement April 23, 2010 Trimming base and preparing for PCC instrumentation placement April 26–30, 2010 PCC placement and instrumentation May 6 and 10, 2010 HMA shoulders May 20, 2010 Open to traffic June 7, 2010 Figure 2.4. Layout of test sections at MnROAD (Cell 70, HMA/PCC; Cell 71, PCC/RCA PCC; Cell 72, PCC/low-cost PCC).

25 Subgrade Soil Grading and Compaction A string line was set for trimming of the subgrade and the base. The subgrade was cut with a trimming machine (Figure 2.9) and compacted with a steel drum roller (Figure 2.10). Hand holes and conduits were set for the instrumentation cables (Figure 2.11). Testing was performed on the compacted sub- grade using a dynamic cone penetrometer (DCP), lightweight deflectometer (LWD), and falling weight deflectometer (FWD). The Class 5 aggregate base was constructed in two 4-in. lifts. PCC Mix Design Numerous options for PCC mixes to be used in the top and lower lifts of the PCC/PCC pavement were explored, which Figure 2.5. Guillotine crusher breaking existing concrete for recycling into the lower PCC layer of the PCC/PCC composite pavement. Figure 2.6. Removal of existing concrete pavement for recycling. the concrete sections before the existing concrete was crushed into coarse aggregate. For this project, the contractor used an industrial crushing operation that included a primary jaw crusher (Figure 2.7) operating at less than full capacity and a secondary cone crusher (Figure 2.8), then washed, screened, and stockpiled. The jaw crusher jaws were distanced to adjust the maximum aggregate size produced. The cone crusher was used as secondary crusher to further remove the mortar from the natural aggregates. A cone crusher squeezes material between an eccentrically gyrating spindle and a bowl below. As the pieces are broken, they fall to the lower, more closely spaced part of the crusher and are fur- ther crushed until small enough to fall through the bottom opening. Laboratory tests on the recycled aggregate (AASHTO T84 and T85) revealed that the RCA absorption was 2.93%. Figure 2.7. The primary crusher was the jaw crusher operating at less than full capacity. Figure 2.8. A cone crusher was used as a secondary crusher to further remove the mortar from the natural aggregates.

26 available natural fine aggregates, and a coarse RCA as a low- cost alternative coarse aggregate. All basic components of the lower-layer PCC were selected in light of a desire to reduce costs, investigate methods of sustainability, and investigate the reuse of materials into structural components. Each of the PCC mixes used is summarized below and in Table 2.3. RCA PCC Mix Design Per the special provisions, the RCA comprised 50% of the total coarse aggregate in the PCC mix. In addition, aggregate fines less than 4.75 mm (No. 4) and coarse aggregates greater than 25.4 mm (1 in.) used in the PCC mix were specified to come from virgin aggregate sources. The special provisions also required the contractor to clean and wash the RCA. As much as 10% of the total recycled coarse aggregate could consist of bituminous particles. The cementitious fraction was specified to consist of as much as 60% SCMs, including but not lim- ited to fly ash. Fly ash replacement of 40% was approved and used in the final mix design. The main concern with regard to this mix had to do with the use of coarse RCA. As a result of these concerns, an extensive investigation into the use of RCA for structural PCC was conducted. This included labo- ratory work investigating aggregate absorption, gradation, freeze–thaw durability, aggregate washing/preparation, and methods of crushing, as detailed in Appendix F. Low-Cost PCC Mix Design Per the special provisions, the cementitious fraction was specified to consist of as much as 60% supplementary Figure 2.9. Trimming the subgrade using a string line and trimmer. Figure 2.10. Compacting the subbase using a steel drum roller. Figure 2.11. Installing conduits to carry and protect the instrumentation cables. involved a series of iterations on mix design, followed by lab- oratory testing of the mixes (see Appendix Q). For this SHRP 2 R21 project, a “high-quality” PCC mix for the upper layer of PCC/PCC composite pavements is defined as a PCC mix containing increased cement content (relative to the American PCC paving standard of roughly 500 to 600 lb/yd3 [297 to 357 kg/m3]) and a high-quality, very durable aggregate (i.e., granite). The aggregate in the upper lift must be gap-graded and of a maximum size no larger than 0.3 in. (8 mm). Although German and Austrian mix designs do not typically contain fly ash, it was used in mixes at MnROAD. In addition, the research team aimed for a PCC mix for the lower PCC that would contain reduced cement (relative to the standard described above), locally

27 cementitious materials including, but not limited to, fly ash. Fly ash replacement of 60% was approved and used in the final mix design for the low-cost PCC mix. The main con- cern with regard to this mix had to do with setting time and early strength because of the high fly ash replacement per- centage. However, it should be noted that although 60% is atypical, this level of cement replacement is possible and has been accomplished for other transportation concretes. Furthermore, a high percentage of cement has successfully been replaced using slag as the SCM in many construction projects throughout Europe. UPPeR LAyeR PCC Mix Design The upper layer PCC mix included high cement content (616 lb/yd3) in addition to 15% fly ash substitution. The mix incorporated polish-resistant, granite aggregates with a high cement content that would allow for an exposed aggregate surface texture. The maximum aggregate size of the coarse aggregate was specified as 9.5 mm. The mix included as-needed hydration stabilizer for slump retention. The portland cement used was a Holcim, St. Genevieve Type 1⁄2 cement. The fly ash was a Class F, Headwaters Coal Creek fly ash. The fine aggregate was an Elk River Concrete Sand. The coarse aggregate comprised No. 67, 3⁄4-in., and No. 4, 11⁄2-in. Elk River gravel. The coarse aggregate for the top layer PCC comprised 3⁄8- and 1⁄2-in. washed granite chips from Martin Marietta. The water reducer, accelerator, and air entrainer were Sika products. The hydration stabilizer was a BASF product. The gravel aggregates and gradation of those aggregates were similar to those of conventional PCC pavements used by Minnesota DOT for the MnROAD facility. PCC Mix Gradation The research team elected to use an EAC surfacing for the demonstration slab and mainline sections. Although EAC is used successfully in Europe, challenges were faced by the Kansas and Michigan DOTs in applying EAC techniques in the United States. A key issue with regard to attaining a low- noise, high-durability EAC texture is the aggregate gradation of the PCC surface mix. Gradations for the PCC surface mix and the lower lift PCC mixes were chosen based on a com- bination of laboratory testing, communication with con- tractors and engineers and review of research reports from construction in Kansas and Michigan, and communication with engineers and contractors in Europe. Ideally, to obtain a high-quality EAC texture, a gap-graded mix (small percent- age between No. 4 and No. 16 sieve sizes) with maximum aggregate size less than 8 mm is desirable for the PCC surface mix. This results in closely spaced aggregates with a negative surface texture. However, limitations of construction funds and sources of aggregates close to MnROAD necessitated modifications to the original specifications. The modifica- tions included use of maximum aggregate size of 9.5 mm and a denser gradation. The gradation ranges specified in the original and updated specifications along with the final approved gradation are shown in Table 2.4. Instrumentation Plan To determine the overall response of the pavement to envi- ronmental loads, the physical response of the pavement and the climatic conditions within the structure were monitored. Envi- ronmental sensors were installed to document the temperature and moisture gradients that developed throughout the depth Table 2.3. PCC Mix Design for PCC/PCC Construction at MnROAD Materials Weight per Cubic Yard (lb/yd3) RCA PCC Low-cost PCC Upper Layer PCC Water 234 173 283 Cement 360 240 616 Fly ash 240 360 109 Sand 1,200 1,263 843 California No. 1 (virgin aggregate, 1½-in. max imum aggregate size) 825 787 na California No. 2 (recycled aggregate) 920 na na California No. 3 (virgin aggregate, ¾-in. max imum aggregate size) na 1,102 na 3/8-in. Washed granite chips na na 843 ½-in. Washed granite chips na na 1,133 Air entrainer 2 to 15 oz 2 to 15 oz 2 to 15 oz Hydration stabilizer na na 0 to 5 oz Water reducer 1 to 5 oz 1 to 5 oz 1 to 5 oz Accelerator 0 to 30 oz 0 to 30 oz na Properties Water-to-cement ratio 0.39 0.29 0.39 Maximum slump 3 in. 3 in. 3 in. Entrained air content 7% 7% 7% Note: na = not applicable.

28Table 2.4. Aggregate Gradation for PCC Mixes Designation 37.5 mm (1½ in.) 31.5 mm (1¼ in.) 25.0 mm (1 in.) 19.0 mm (¾ in.) 16.0 mm (3/8 in.) 12.7 mm (½ in.) 9.5 mm (3/8 in.) 6.3 mm (¼ in.) 4.75 mm (No. 4) 2.36 mm (No. 8) 1.18 mm (No. 16) .030 mm (No. 50) .015 mm (No. 100) .0075 mm (No. 200) Working range limits ±5 ±5 ±5 ±5 ±5 ±5 ±5 ±5 ±5 ±4 ±4 ±3 ±2 1.6% maximum RCA PCC Mix As-written specifications 100 97–87 87–77 76–66 70–60 63–53 55–45 na 41–31 30–20 23–13 14–4 10–1 7–1 Updated specifications 100 100 95–80 85–70 na 70–55 60–45 55–40 50–35 45–30 35–25 10–2 10–0 5–0 Final approved PCC blend 100 100 88 73 na 54 46 na 41 37 30 6 1 0.2 Low-cost PCC Mix As-written specifications 100 97–87 87–77 76–66 70–60 63–53 55–45 na 41–31 30–20 23–13 14–4 10–1 7–1 Updated specifications 100 100 95–80 85–70 na 70–55 60–45 55–40 50–35 45–30 35–25 10–2 10–0 5–0 Final approved PCC blend 100 100 89 76 na 64 56 49 42 37 30 6 1 0.2 Upper Layer High-quality PCC Mix As-written specifications 100 100 100 100 100 100 100 75–65 48–38 48–38 48–38 13–7 7–1 5–1 Updated specifications 100 100 100 100 100 100 100–95 75–65 55–45 40–30 35–25 13–7 7–1 5–0 Final approved PCC blend 100 100 100 100 100 100 98 69 48 33 29 11 2 0.4 Note: na = not applicable.

29 of the slab. Temperature sensors were located in each of the different pavement structures so that the seasonal, daily, and construction temperature profiles that developed could be documented. Moisture sensors were installed in the concrete to study the effects of the surface layers on the moisture dis- tribution through the depth of the slab. Static strain gauges were used to monitor the effects of uniform moisture and temperature changes, as well as moisture and temperature gradients on the slab shape. Figure 2.12 shows the elevation and plan view of the instrumentation layout. The response of the structures to applied vehicle loads was measured using dynamic strain sensors installed within the pavement structure. An on-site weather station recorded air temperature, relative humidity, and wind speed every 15 minutes. The various sensors installed at the MnROAD test section are described here: • Temperature sensors: Thermocouples were used for mea- suring temperature throughout the pavement structure. Critical locations for monitoring temperature included the midslab, the slab corner, and midslab adjacent to the longitudinal joint. • Concrete moisture: To measure moisture levels within the concrete, 24 Sensirion SHT75 relative humidity and temperature sensors were installed. The SHT75 sensor is a relatively small (approximately 0.75 × 0.25 × 0.125 in.) and cost-effective means of measuring relative humidity in concrete. • Static strain: The PCC response to static loads generated was measured with vibrating wire (VW) strain gauges. The VW gauges were used to provide several critical pieces of information related to the performance of the PCC layers, including: 4 Degree of bonding between the PCC layers; 4 Slab curvature; and 4 In-place drying shrinkage and thermal coefficient of expansion. Geokon Model 4200 VW concrete embedment strain gauges were used. The gauges operate on the VW principle. A steel cable is tensioned between two metal end blocks. When the gauge is embedded in concrete and concrete deformations occur, these end blocks move relative to one another. The movement of these end blocks influences the degree of tension in the steel cable. The tension in the cable is quantified by an electromagnetic coil, which mea- sures the cable’s resonant frequency of vibration on being plucked. The sensor is also equipped with a thermistor so corrections for temperature can be made. • Dynamic strain: Dynamic strain sensors were installed to measure the pavement response to loads applied by truck traffic and the FWD. The dynamic sensors used in the con- crete were Tokyo Sokki PML-60-2L strain gauges. The Tokyo Sokki PML-60-2L consists of a copper/nickel alloy resistance foil gauge attached to two lead wires. This foil is attached to an electrically insulated backing and, with the use of a special adhesive, is attached to one of two thin acrylic plates. The two plates are sealed together to protect the gauge from contamination when installed in the concrete. These acrylic plates are coated with a fine, granular material to improve bonding to the surrounding concrete. The insulated backing expands and contracts with the concrete, causing the resis- tance in the foil gauge to change. • Data acquisition: Automated static and dynamic data (or “online” data) were entered into the MnROAD database through the MEGADAC acquisition system. This system of dynamic cabinets, computers, fiber-optic cables, and copper-wire sensors automatically retrieved data from instruments at the MnROAD facility in Albertville and returned this information to the MnROAD database in Maplewood. Instrumentation Installation Figure 2.13 shows dynamic strain gauges, static strain gauges, humidity sensors, and thermocouples affixed to the aggregate base prior to PCC placement. The lead wires are buried in the sublayers and carry the signal from the gauges to the data acquisition unit. The gauges were packed in the concrete, and the concrete was vibrated with a hand vibra- tor to ensure consolidation of concrete around the gauges (Figure 2.14). Paving Operations Paving operations for the R21 test sections at MnROAD began on April 28, 2010, with the construction of a 200-ft demonstra- tion slab, and concluded on May 10, 2010, with the completion of 950 ft of test sections along the mainline (I-94) test area. The two-lift paving used two GOMACO model GHP2800 pavers and a belt placer between the two pavers to place fresh mix for the upper lift (Figure 2.15). Many aspects of the paving were similar to those of a normal single-layer PCC pavement. As detailed in Table 2.1, the pavement design included 1.25-in. dowels, placed at the middepth of the full PCC slab using dowel baskets. Further- more, the design included 30-in., no. 4 tie bars spaced at 30 in. to reinforce longitudinal joints; the bars were inserted using a tie-bar inserter attachment on the first paver. One difference in the use of two pavers in PCC/PCC versus single-layer PCC is that the upper lift paver was adjusted to “crown” the lower lift slab by 0.75 in. on each side; that is, the second paver paved a lift 1.5 in. wider than the first paver in the train.

30 .5" 1.5" 2.5" 3.25" 4" 6" 8" 9" 13" 17" 20" 26" 6" PCC Special 2 8" Class 5 Existing Clay Subgrade Wheelpath Legend Static Strain Gauge Thermocouple Sensor Fence Centerline Riser BD Pedestal Conduit Dynamic Strain Gauge Moisture Gauge Thermocouple TreeHand hole 12' 10' Bit Shoulder 15' 15' 15' 15' 15' Static PanelsDynamic Panels Spacing Panel- No Sensors ~45' P1P2P3P4P5 3" 2" 3" 2" 4" 4" New Cab 2-4" Passing Lane Driving Lane N 3" PCC Special 1 Figure 2.12. Elevation and plan view of instrumentation layout for PCC/PCC test sections at MnROAD.

31 Figure 2.13. Instrumentation installed prior to placement of the PCC to measure pavement responses to temperature and traffic loads (static strain gauge, top left; dynamic strain gauges, bottom left; humidity sensors, top right; temperature sensors [thermocouple tree], bottom right). Figure 2.14. Overview of instrumentation packed in concrete before PCC paving.

32 Figure 2.15. Paving train constructing R21 test sections along I-94 at MnROAD (from left to right: the mixer truck, first paver, belt placer, and second paver). Unlike the paving trains encountered in Europe, which use low-frequency vibration on the second paver only, the paving train used for the MnROAD construction used vibra- tors in both pavers (the first paver set to 8,000 vpm, the sec- ond paver set to 4,000 vpm). Methods differ on this point, in part because of the use of automated dowel bar inserters in Europe. Because of the use of dowel baskets for placing the dowel bars, vibrating the lower lift PCC was necessary to consolidate the PCC mix around the dowel bars. However, given the height of the dowel baskets (4.5 in.) and the small thickness of the lower PCC lift (6 in.), the vibrations were surficial. The vibration in the second paver was low and shal- low to avoid overmixing the two PCC layers, particularly at the interface, and thus ensure the integrity of the individual layers. The key complications with respect to the paving were those brought about by delays in the delivery of PCC for the two lifts. Although the construction specifications indi- cated that paving of the second lift was to occur no later than 90 minutes after the first lift (ideally no later than 60 minutes), on all three occasions of PCC/PCC paving (demonstration slab and two mainline sections), the paving was frequently stalled for more than 90 minutes as crews waited for batched upper lift PCC to arrive. During the construction of the demonstration slab, mix delivery delays led to 90- to 100-ft stretches of the placed lower lift being exposed to the envi- ronment for more than 120 minutes before the second lift was placed. These delays resulted in a few problems that could be observed immediately on-site during paving. The most appar- ent was the setting up of concrete in the auger, the grout box, and on the profile pan of the paver. Frequent delays allowed the concrete to hydrate and attach to surfaces, normally assumed to be smooth, that physically form the slab. When paving resumed after long delays, concrete that had clung to these surfaces would “tear” at the freshly paved concrete, resulting in the need for additional finishing. Figure 2.16 illus- trates the tearing. The delays in the delivery of the upper PCC compro- mised the pavement, given that the weather during the demonstration slab paving was unseasonably warm, sunny, and windy. Temperatures were between 60°F and 69°F, the sun was strong with no clouds, and the wind was steady at 5 to 10 mph with occasional strong gusts. These conditions are especially critical when the slab in question was composed of the early batches of PCC that arrived for the demonstra- tion slab, which were considerably dry (with measured slump on-site of 0.75 to 1 in. from batch to batch). This early dry PCC was used for the 90 to 100 ft of lower lift placed at the beginning of the demonstration slab, for which more than 120 minutes passed before an upper lift was placed. Figure 2.16 illustrates the most exaggerated of the shrinkage cracking encountered in these early slabs. Figure 2.16. At left, “torn” edges and surface caused by concrete setting on various parts of the paver. At right, coring showed poor mix consolidation in the lower PCC at select locations.

33 Another concern about delays included the integrity of the bond at the interface of the two lifts. An ultrasonic tomogra- phy testing device was used to assess the bond at the trans- verse joints and at midslab locations on the demonstration slab. Tomograms from two representative scans are illus- trated in Figure 2.17. Ultrasonic reflection occurs only noticeably at the start of the base layer (measured as approximately 8 in.) in the tomo- gram at left in Figure 2.17. In the tomogram at right, however, significant ultrasonic reflections are measured at a depth of approximately 4 in., near the interface of the two PCC layers. This reflection near the interface may be indicative of a poorly developed bond between the two layers of PCC, or it may be indicative of other problems (such as tearing and voids) in the pavement. The figure at left is further evi- dence that the composite layers, from the view of the tomo- gram, are a unified layer, whereas the reflections in the figure at right suggests the possibility of internal distress (these conclusions were confirmed by cores taken from the dem- onstration slab). Many lessons (with regard to mix delivery, lower PCC slump, instrumentation installation, saw cutting, and so forth) were learned from the construction of the demonstration slab that was incorporated during the construction of the mainline test sections. However, despite the problems encountered during the construction of the demonstration slab, a large portion of the PCC/PCC constructed (particularly the second 100 ft) was found (based on coring and trenching evidence) to meet the requirements of the design with good integrity of the individual PCC layers (no intermixing) and good bond between the PCC layers (Figure 2.18). Figure 2.17. At left, a typical tomogram from the PCC/PCC demonstration slab at MnROAD; at right, tomogram with ultrasonic reflection near the depth of the PCC-PCC interface. 3-in EAC 6-in RCA 8-in Aggregate Base Figure 2.18. Cross section of the PCC/PCC demonstration slab constructed at MnROAD showing very good integrity and bond between the two PCC layers.

34 Paving the mainline sections progressed at a rate of between 1 and 4 ft per minute, and the project contractor was confi- dent that this rate would be greatly increased with a larger project and a consistent supply of PCC for paving. While joint cuts sawed to a depth of 3 in. did not necessarily propagate well on the demonstration slab because of construction delays and dry mixes, for the mainline sections all saw cuts were found to propagate as anticipated for a single-lift equivalent slab. The specifications were changed to saw cut one-third of the total thickness or top layer thickness plus 0.5 in., whichever is greater. In this case, it was a minimum of 3.5 in. Mix Design and Delivery One of the more challenging aspects of the PCC/PCC sections constructed at MnROAD was the PCC itself. This challenge presented itself in: (1) the development of a mix design that uses alternative materials and/or meets low-cost specifications and (2) terms of the logistics behind batching and delivering concrete to meet the demands of the paving operations. The characteristics of the three PCC mixes used are sum- marized in Tables 2.3 and 2.4. The most conventional of the three is the PCC mix used for the upper lift, whereas the PCC used for the lower lifts presented challenges in its use of high fractions of fly ash and/or RCA. The specification for as much as 60% fly ash in the lower lift PCC was inspired by the high-fraction of SCM replacement in the new St. Anthony Falls (I-35W) bridge in Minneapolis, which used as much as 81% SCM replacement in its mixes. The existence of a lower lift was also viewed as an opportunity to use lesser qual- ity aggregates. To this extent, a thorough review of existing research on the use of RCA in PCC was performed. This review concluded that RCA was a viable coarse aggregate for the lower lift PCC provided the RCA came from a known source, fines were excluded, and the stockpile was properly maintained (i.e., kept saturated to eliminate variable absorp- tion as a concern). Both the use of high-fraction SCM replacement and RCA came with the challenges discussed above. These challenges were met with mixing preliminary batches of each mix in the laboratory (Figure 2.19). As a result of these tests, adjust- ments to the mixes were made. Although this preliminary work addressed some chal- lenges, the two-lift paving at MnROAD revealed a larger problem for the concrete in terms of consistency from batch to batch. The challenge of providing a consistent batch from truck to truck was thought to be overcome after the demon- stration slab. However, paving on the mainline again suffered from the consistency problem, particularly in the case of the lower PCC mixes, whose as-delivered slump varied between 0.25 and 2.75 in (the target slump was 1 in.). Although the causes of this inconsistency are still uncertain, there are numerous possible causes: • The use of RCA requires close attention. The contractor had secured RCA of a known source and had washed the RCA of fines; however, the preparation of the RCA for batching, most notably, its degree of saturation, was not consistent. One explanation of the inconsistency from batch to batch, as evident in the variable slump, is the inadequate mainte- nance of the RCA stockpile (Figure 2.20). It is possible that portions of the stockpile had been allowed to dry. • Another concern to emerge from the use of RCA was the underestimate of unprocessed recycled concrete required Figure 2.19. At left, PCC specimens cast in preparation for the paving and brushing of the PCC/PCC demonstration slab; at right, top-surface PCC mix being placed in front of the second paver on-site at MnROAD.

35 to achieve a coarse aggregate of a desired size. Early esti- mates missed the amount of recycled concrete required, which led to only 275 ft of PCC/PCC using the RCA mix being paved, instead of the originally planned 475 ft. Note that RCA was used in construction of Cell 70, the HMA/ PCC test section, and the PCC/PCC demonstration slabs and several truckloads had to be rejected (for a variety of reasons, including slump and entrained air). • The ready-mix supplier used by the contractor did not fre- quently design concretes using a large fraction of fly ash. As a result, the ready-mix supplier’s inexperience in fly ash led to the mix designs being inadequately composed to handle such large amounts of this SCM (in terms of water demands, admixtures, and so forth). • A final challenge in meeting the mix design for the PCC/ PCC pavements was the use of a local ready-mix supplier for the PCC/PCC concrete. Because of the small size and scope of the project, the contractor used a local ready-mix plant, instead of a mobile batching plant, and the use of one plant instead of two (as observed in Europe). Thus, the observed delays in mix deliveries may have been attributable to the use of a ready-mix supplier that was inexpe- rienced in certain mix designs and in delivering those designs in sufficiently large volumes. During the post-construction review, the contractor maintained that one plant was enough to accommodate the three mixes for this project, but the con- tractor also stated that a ready-mix plant was not sufficient to provide consistency in mix design and delivery. The contractor was confident that for a larger project, using the company’s own mobile batching plant and staff (rather than subcontract- ing this work to a local ready-mix supplier), mix consistency/ delivery would not complicate PCC/PCC paving. Figure 2.21 shows the placement of the lower PCC mix (note the dryness and low slump of the mix, which was specified at 1 in.). Figure 2.22 shows the placement of the upper PCC mix above the stiff lower PCC layer using a belt placer. The lower PCC layer was stiff enough to carry the impact and weight of the upper PCC layer while being placed (and the weight of an average size person as seen by the footprint impressions on the lower PCC). The footprints also show that the lower PCC layer was still “wet,” which is necessary for a good bond between the two PCC layers. Surface Texturing The finishing platform used for the construction was a GOMACO model TC600 with Power Pavers Inc TC 2700T spray attachment. After paving, a curing/retarder compound (MBT Reveal from BASF Building Systems) was applied to the surface that both acted as a moisture barrier (curing agent) and as a retarder of hydration in the PCC surface. During construction of the demonstration slabs, early applications Figure 2.20. RCA stockpile being processed and saturated at a concrete recycling facility. Figure 2.21. Placement of the lower PCC mix. Figure 2.22. Placement of the upper PCC mix on the lower PCC layer.

36 of the surface treatment were delayed because of mechanical problems on the finishing platform, which provided insuffi- cient pressure to the spray nozzles. The treatment (Figure 2.23) was intended to be applied almost immediately after finishing of the placed second lift; however, because of frequent delays, the treatment was applied anywhere between 60 and 90 min- utes after the completion of paving a given segment. For the construction of the mainline sections, the nozzle heights were adjusted and wind guards were attached to the side of the cur- ing cart to apply the compound more uniformly. For the demonstration slab and the first day of mainline paving, brushing was initiated anywhere between 5 and 8 hours after paving of a given section had completed. The brush timing was based on limited laboratory tests, which did not mimic field conditions closely. To compensate for the lack of field experience, the surface was frequently tested at regular inter- vals, judging the brush readiness of the surface by the amount of cement and aggregate dislodged using a metal rod and/or handheld brush. The brushing was accomplished using a small front-end loader with rotating wire brush attachment (Figure 2.23). The brushing was complicated by the inability of the operator to know the depth of texturing with any kind of precision. Thus, the brushing was done in multiple passes to gauge the level of cement removal between the aggregates, slowly revealing the EAC texture in pass after pass (Figure 2.24). The extent Figure 2.23. Treated surface and finishing platform (left). Equipment for EAC brushing (right). Figure 2.24. Surface after first pass with brush (left). Finished EAC surface (after wash) (right).

37 of brushing was determined using a combination of a sand patch test and an aggregate peak counting test (Figure 2.25). More detail on these techniques can be found in Weinfurter et al. 1994. Although it was not specified, the aggregate peak counting test was an informal quality control for the brush- ing, adapted from Austrian methods. It aimed for a count of anywhere between 40 and 50 aggregate points per 25 cm2 (3.88 in.2), according to Haider et al (2006). The sand patch test was conducted according to ASTM E965 at intervals as a quality control measure during brushing to ensure that the mean texture depth (MTD) was between 0.8 and 1.2 mm (0.03 to 0.05 in.), as specified. This target was based in part on German and Austrian specifications for texture depth (0.6 to 0.8 mm and 0.8 to 1.0 mm, respectively). The completed average MTD for the EAC surface was 0.76 mm (0.030 in.). Two to five passes were needed to obtain the desired texture. During the paving of the second PCC/PCC test section on the mainline, the construction encountered sudden onset of rain in the late afternoon. A vast portion of the finished, treated PCC/PCC paved was subjected to the rain before being covered with polyurethane sheeting. The delay in sheet- ing was caused by delays in paving and then in the application of the surface treatment. For these reasons, the brush tim- ing was uncertain, and brushing was not initiated until the morning of the next day, 20 hours after the second lift had been placed and after the joints had been sawed. Although an EAC texture was still obtained, because of the various fac- tors (localized washing of some of the curing/retarding com- pound by rain, low temperatures, variability in application of the curing compound, and so forth) the final texture was less uniform, with areas of good EAC texture and other areas of insufficient exposed aggregate. The situation was a reminder of the need to remain aware of the weather and sheet the PCC as soon after placement as possible should rain occur. As shown in Figure 2.26 and described, the initial 475-ft portion of the PCC/PCC EAC texture (all of Cell 71 and a portion of Cell 72) was diamond ground, resulting in a total of three surface textures for the PCC/PCC sections: • 475-ft passing and driving lane EAC; • 475-ft driving lane next-generation diamond grind; and • 475-ft passing lane conventional diamond grind. As-constructed Properties The FHWA Mobile Concrete Laboratory visited the R21 MnROAD construction site and collected PCC cores and material samples. The results, which are the average of two tests, are summarized in Table 2.5. According to the Materials and Construction Optimiza- tion project (National Concrete Pavement Technology Cen- ter 2008), for adequate protection of concrete in freeze–thaw environment, spacing factor values less than 0.01 in. are desirable, although values less than 0.015 in. are commonly considered acceptable. The spacing factors of all the samples from this project were less than 0.015 in. In fact, the spacing factors of three of the four samples were less than 0.01 in. For specific surface, which indicates the size of the air bub- bles, values greater than 600 in.–1 are desirable. Three of the four samples that were tested had values higher than 600 in.1. Based on the test results for specific surface and spacing fac- tor, the mixtures used in the composite pavements project Figure 2.25. Quality control tests for brushing: (left) 25 cm2 test to count aggregate peaks and (right) sand patch test to determine texture depth.

38 have good air void distribution for protection against freeze– thaw damage. Noise Measurements Construction of the EAC finish was attempted because of its durability and because it channels water away from the wheel path in multiple directions. However, the primary benefit of a properly constructed EAC surface is its noise mitigation potential. On-board sound intensity (OBSI) measurements of all of the finished composite pavement surfaces were col- lected to compare the sound intensity of the various surface finishes (Akkari and Izevbekhai 2011). Noise data from the EAC and diamond ground surfaces were compared with those from the HMA/PCC composite pavement. The OBSI test setup consists of a sedan outfitted with four GRAS sound intensity meters, a Brüel & Kjær front-end four- channel frequency analyzer, and a standard reference test tire (SRTT). The microphones are suspended from the vehicle frame and positioned at 3-in. vertical displacement and at 2-in. lateral displacement from the leading and trailing end of the standard reference tire and pavement contact. The microphones are anchored to a free rotating ring mounted on the right wheel that allows the microphone assembly to be fixed in position and direction without inhibiting the rotation of the tire. The OBSI equipment is shown mounted to a sedan wheel in Figure 2.27. PULSE noise-and-vibration software is installed in a con- nected computer. The computer receives and analyzes the data, categorizing the response into component third octave frequency output. Pavement noise response from the micro- phones is condensed into a third octave frequency sound intensity plot averaged for the leading edge and trailing edge. The OBSI parameter is the average of the logarithmic sum of the sound intensity at 12 frequencies (400, 500, 630, 800, 1,000, 1,250, 1,600, 2,000, 2,500, 3,150, 4,000, and 5,000 Hz). OBSI analysis is based on the AASHTO TP76-08 protocol. The results from OBSI testing done in 2010 are shown in Fig- ure 2.28. Figure 2.28 shows that the innovative diamond-ground finish (also called the next-generation) had the lowest OBSI throughout the 3 months tested. The traditional diamond grind had an OBSI similar to that of the hot-mix asphalt (HMA) surface. The EAC surface had the highest OBSI. There was not a considerable difference between the OBSI in the passing lane compared with that in the inside lane in either Cell 70 or 72. In a survey of exposed aggregate con- crete pavements in Europe conducted by the National Con- crete Pavement Technology Center, OBSI values were found to range from 101 to 106 dBa, which is similar to the results Next-GenerationConventional Figure 2.26. The initial learning portion of the EAC texture was diamond ground using conventional grinding in the passing lane and next-generation grinding in the driving lane.

39 Figure 2.27. OBSI device. obtained for Cell 72. The one-third octave sound-intensity spectrums used to calculate OBSI values are shown in Fig- ure 2.29; great similarities exist in the spectrums of the vari- ous surfaces, except for that of the IG surface. Field Survey Sections Introduction In addition to the constructed test sites, additional field sites were identified to cover other types of PCC/PCC com- posite sections and to bring some long-term performance data into the analysis. An on-site condition survey was conducted, and detailed information regarding traffic, materials, and additional performance data was collected from the highway agencies. Table 2.6 provides a list of the PCC/PCC composite pavements that were included in the database, along with key design and construction informa- tion and photographs. Table 2.5. As-constructed PCC Mix Propertiesa Property Cell 71 RCA Mix Cell 71 Surface PCC Mix Cell 72 Low-cost Mix Cell 72 Surface PCC Mix Entrained air content, % 6.5 4.5 6.5 6.5 Unit weight, lb/ft3 145.3 145.12 148.4 142.8 Flexural strength, psi (7 day) 527 606 468 790 Flexural strength, psi (14 day) 578 798 515 897 Flexural strength, psi (28 day) 665 891 548 816 Flexural strength, psi (90 day) 785 958 669 1011 Compressive strength, psi (7 day) 3,599 5,314 3,618 5,289 Compressive strength, psi (14 day) 3,890 5,628 4,071 5,260 Compressive strength, psi (28 day) 4,305 5,855 5,062 5,663 Compressive strength, psi (90 day) 5,663 7,598 6,695 7,388 Modulus of elasticity, psi (7 day) 4.30 × 106 4.76 × 106 4.73 × 106 4.45 × 106 Modulus of elasticity, psi (14 day) 4.87 × 106 4.82 × 106 NA 4.44 × 106 Modulus of elasticity, psi (28 day) 4.83 × 106 4.95 × 106 5.11 × 106 4.83 × 106 Modulus of elasticity, psi (90 day) 5.42 × 106 5.46 × 106 5.77 × 106 5.04 × 106 Poisson’s ratio (7 day) 0.22 0.26 0.21 0.25 Poisson’s ratio (14 day) 0.25 0.24 NA 0.22 Poisson’s ratio (28 day) 0.25 0.26 0.23 0.21 Poisson’s ratio (90 day) 0.30 0.28 0.23 0.23 Split tensile strength, psi (28 day) 337 392 343 390 °F 5.8 × 10-6 5.6 × 10-6 5.4 × 10-6 5.6 × 10-6 Air void analyzer (AVA) spacing factor, in. 0.0088 0.0125 0.0088 0.0082 AVA specific surface, in.-1 580 639 628 710 Note: NA = not available. a FHWA Mobile Concrete Laboratory.

40 Source: Akkari and Izevbekhai 2011. O BS I (d BA ) Figure 2.28. OBSI for the HMA, innovative diamond-ground, traditional diamond-ground, and EAC finishes. Source: Akkari and Izevbekhai 2011. Frequency (Hz) So un d In te ns ity L ev el (d BA ) Figure 2.29. The one-third octave sound intensity spectrum for composite pavement surfaces on July 8, 2010.

41 Table 2.6. PCC/PCC Composite Pavement Field Sections Composite Pavement Type Location Construction Year and Traffic Comments 3-in. PCC over 9-in. econocrete (low- quality PCC with fc = 2,000, 1,250, or 750 psi) Nondowel, 15-ft skew joints Fort Myers, Florida US-41 12 Test sections 1978 New construction 11,000 average daily traffic (ADT) (12% trucks) 1992: Rigid pavement performance report does not indicate perfor- mance problems. 2010: Greene et al. note good bond and performance over 30-year life. 3-in. PCC over 9-in. econocrete (low- quality PCC with fc = 2,000, 1,250, or 750 psi) Nondowel, 15-ft square joints Fort Myers, Florida US-41 12 Test sections 1978 New construction 11,000 ADT (12% trucks) 1992: Rigid pavement performance report does not indicate perfor- mance problems. 2010: Greene et al. note good bond and performance over 30-year life. 3-in. PCC over 9-in. econocrete (low- quality PCC) with fc = 2,000, 1,250, or 750 psi 1-in. dowel, 20-ft square joints Fort Myers, Florida US-41 12 Test sections 1978 New construction 11,000 ADT (12% trucks) 1992: Rigid pavement performance report does not indicate perfor- mance problems. 2010: Greene et al. note good bond and performance over 30-year life. From Greene et al. 2010 From Greene et al. 2010 From Greene et al. 2010 From Greene et al. 2010 (continued on next page)

42 Table 2.6. PCC/PCC Composite Pavement Field Sections Composite Pavement Type Location Construction Year and Traffic Comments 3-in. PCC over 7-in. PCC (including slightly reduced cement content) 20-ft joint spacing Rock Rapids, Iowa US-75 1976 New construction Traffic unknown Estimated 6000 ADT (13% trucks) 2004: Cable and Frentress indicate pavement performing well to date. 3-in. alkali-silica reactivity (ASR)- susceptible PCC with lower water-cement ratio over 7-in. high absorption limestone PCC 15-ft joint spacing Section 12 Haven, Kansas K-96 1997 New construction 4,800 ADT (11% trucks) 1998: Wojakowski reports no initial performance concerns. 2011: Some evidence of ASR, but overall the pavement performing exceptionally well with no other distresses. 3-in. ASR-susceptible PCC with 20% pozzolan (Dura-Poz) replacement for ASR mitigation over 7-in. high- absorption limestone PCC 15-ft joint spacing Section 11 Haven, Kansas K-96 1997 New construction 4,800 ADT (11% trucks) 1998: Wojakowski reports no initial performance concerns. 2011: Some evidence of surface edge tearing/cracking (not fatigue) but overall the pavement perform- ing exceptionally well with no other distresses. No evidence of ASR. (continued) (continued on next page)

43 Table 2.6. PCC/PCC Composite Pavement Field Sections Composite Pavement Type Location Construction Year and Traffic Comments 3-in. normal PCC over 7-in. PCC with 15% RAP 15-ft joint spacing Section 9 Haven, Kansas K-96 1997 New construction 4,800 ADT (11% trucks) 1998: Wojakowski reports no initial performance concerns. 2011: No major distresses. Only two locations of minor spalling that have been covered with HMA near the longitudinal joint. 2.5-in. PCC over 7.5-in. PCC 15-ft joint spacing Detroit, Michigan I-75 1993 New construction Traffic unknown, estimated 120,000 ADT (13% trucks) 1996: Smiley details early noise prob- lems and localized distress, otherwise performing well. 2010: No longitudinal or transverse fatigue cracking. Wheelpath exhib- its severe spalling, some of which has been patched. Spalling appears to originate from deterioration of the lower PCC at the joints. EAC texture looks good. (continued on next page) (continued)

44 Table 2.6. PCC/PCC Composite Pavement Field Sections Composite Pavement Type Location Construction Year and Traffic Comments 1.5-in. PCC over 11.8-in. PCC 15-ft joint spacing Eight different surface textures Salina, Kansas I-70 2008 New construction Opened December 2008 13,000 ADT (31% trucks) 2011: Surveyed. No structural problems observed. Pavements performing well. Some popouts of longitudinal groove texture and minor surface edge tearing/cracking (not fatigue) observed at some loca- tions. Minor ASR observed. (continued) (continued on next page)

45 Table 2.6. PCC/PCC Composite Pavement Field Sections Composite Pavement Type Location Construction Year and Traffic Comments 3-in. PCC over 6-in. PCC (including RCA) 15-ft joint spacing MnROAD, Albert- ville, Minnesota I-94 2010 New construc- tion 25,000 ADT (14% trucks) 2011: No performance issues. Section performing well with no distresses. Excellent bond between PCC layers. 3-in. PCC over 6-in. PCC (low-cost) 15-ft joint spacing MnROAD, Albert- ville, Minnesota I-94 2010 New construc- tion 25,000 ADT (14% trucks) 2011: No performance issues. Section performing well with no distresses. Excellent bond between PCC layers. 2-in. PCC over 7.9-in. PCC (including RCA) 18-ft joint spacing Traun, Austria A1 1994 New construction 110,000 ADT (13% trucks) 2010: R21 tour noted impressive structural performance and little wear in wheelpath. No distress existed after 16 years. 1.6-in. PCC over 8.3-in. PCC (including RCA) 18-ft joint spacing Eugendorf, Austria A1 1993 New construction 56,000 ADT (13% trucks) 2010: R21 tour noted impressive struc- tural performance (no cracks), yet signs of wear in wheelpath resulting from tire chains after 17 years. (continued on next page) (continued)

46 Table 2.6. PCC/PCC Composite Pavement Field Sections Composite Pavement Type Location Construction Year and Traffic Comments 3-in. PCC over 7.5-in. PCC 15-ft joint spacing Montreal, Quebec, Canada A15 2009 New construction Traffic unknown, estimated 150,000 ADT (13% trucks) 2011: Ministry of Transportation of Quebec indicates sections performing well with no major distresses. 2.8-in. PCC over 7.5-in. PCC 18-ft joint spacing Bavaria, Germany A93 1995 New construction 70,000 ADT (25% trucks) 2010: Test sections developed to investigate texturing (including EAC), overall impressive structural performance of sections given heavy traffic after 15 years. 3.5-in. PCC over 7-in. PCC 18-ft joint spacing Veghel, the Netherlands N279 2000 New construction Traffic unknown, estimated 25,000 ADT (30% trucks) 2010: Test sections developed to investigate EAC texture depth. R21 tour noted good structural performance after 10 years. Note: fc = 28-day compressive strength. (continued)