U.S. Department of Transportation
Federal Highway Administration
1200 New Jersey Avenue, SE
Washington, DC 20590
|Previous||Table of Contents||Next|
A series of field tests were carried out to assess the placement and condition of the HMA overlay along route STH 55 with and without the use of the Safety EdgeSM device. The objective or purpose of this field study was to evaluate the quality of the in-place HMA material and Safety EdgeSM by investigating three issues or features.
This project was located in Menominee County on STH 55 from the intersection with STH 47 near Keshena and extending north 18.5 mi (project stationing 100+50 to 1080+70). The location of the project is shown in Figure 1. The maximum posted speed limit was 55 mph. The contractor was Northeast Asphalt (NEA).
Figure 1. Project location.
The existing pavement was two lanes of 4-inch HMA over a crushed aggregate base. The lane width was 11 ft for the existing roadway and was planned to be 11-ft wide after construction except in a few isolated areas the pavement was planned to be widened to 12-ft lanes. The aggregate shoulder varied from 1 to 4-ft wide for both the existing and new pavement. New construction consisted of milling the existing pavement 0.5 inches deep and placing a 2-inch HMA overlay. The asphalt mix design was a 12.5 mm Superpave Nd 40 design and included RAP and recycled asphalt shingles. WisDOT has experience with using recycled asphalt shingles. Plans for the shoulder specified a 0.75-inch nominal size dense graded aggregate placed 1.5 inches thick, the contractor planned to use clean pit run crushed gravel. A schematic of pavement cross sections is shown in Figure 2.
This overlay project included several long tangent sections and well shaped shoulders, suitable for demonstrating the Safety EdgeSM which was built on both the northbound and southbound lanes for the length of the project. The slope was designed to be 30°. Details or drawings for the construction of the Safety EdgeSM were not included in the plans. The Safety EdgeSM specification was included in the contract via addendum.
Figure 2. Pavement cross sections.
The northbound lane and about a quarter of the southbound lane had been paved with the TransTech device prior to the site visit. Rain delays during the site visit limited field observations to only a few hours during which time the contractor utilized the TransTech and the Carlson Prototype #2 devices. The Carlson Prototype #3 device was used several days later.
Three Safety EdgeSM test sections and one non-Safety EdgeSM control section were established in the southbound lane approximately 11.5 mi north of Spirit Rock. Spirit Rock is a well known cultural landmark on STH 55. The following summarizes the pavement sections:
Slope measurements were recorded on test section #1, #2, and #3 at 25-ft intervals using a straight-edge and ruler to measure the horizontal and vertical dimensions of the Safety EdgeSM as shown in Figure 3. Slope measurements are listed in Table 1 (all tables are located at the end of this report). The following summaries the average slope measurements.
|Test Section #1||35°|
|Test Section #2||33°|
|Test Section #3||36°|
Figure 3. Slope measurement technique.
Accurate Safety EdgeSM thickness measurements were not possible due the new overlay extending over the edge of the existing pavement and exaggerating the edge thickness.
Three pairs of cores were cut from each test section. The laboratory-determined densities from these cores serve to calibrate the nuclear density measurements. Laboratory density was determined from the bulk specific gravity at saturated surface dry test condition. Each pair of cores were taken from the center of the mat where the maximum number of roller passes occurred and adjacent to the edge where fewer roller passes were made. Tables 2 and 3 include a summary of these test results; core thickness and bulk specific gravities (saturated surface dry) converted to bulk densities.
Figure 4 shows a comparison of the core densities taken along the edge and near the center of the lane for the Safety EdgeSM and control sections. As expected, the densities near the center of lane are significantly higher than along the edge of the mat.
Figure 4. Comparison of core densities adjacent to the edge and at the center of the lane.
Density tests were conducted using a nuclear density gauge in backscatter mode for 60 second test durations. The tests were conducted adjacent to the edge and at the center of the lane at 50-ft intervals and at the location of each core. Figure 5 shows a comparison of the nuclear densities and densities measured on the cores. As shown, there is close correlation between the nuclear and core densities.
Figure 5. Comparison of the nuclear densities and core densities.
Adjustment factors were determined from correlating the nuclear density readings and the core laboratory density results shown in Table 3. The factors were used to adjust the nuclear density gauge readings to be consistent with the densities that were measured in the laboratory. The following summarizes the adjustment factors determined for this project.
|Adjacent to the edge||1.013|
|Center of lane||1.003|
As shown, the value near the center of the lane is closer to unity than the value near the edge. The adjusted or corrected densities using the correction factors are also listed in Table 4.
As expected, the results of the nuclear density tests of each test section show the densities adjacent to the edge were lower than the densities at the center of the lane. The control section had a slightly higher average density value (137.1 pcf) adjacent to the edge compared to the average density of all the Safety EdgeSM sections (134.9 pcf) even though the rolling pattern was assumed to have been the same for all the sections. Test sections #1 and #2 were observed as receiving identical rolling patterns and based on discussions with the contractor, the rolling pattern would have been the same for test sections #3 and #4. The paving of the latter two test sections occurred after the site visit and was not directly observed.
Figure 6 shows a comparison of the nuclear densities taken adjacent to the edge and at the center of the lane. Figure 7 compares the air voids (as calculated from the density test results and the maximum theoretical mix density). The two figures show the densities were lower and the air voids were higher adjacent to the edge than away from the edge.
Figure 6. Comparison of the nuclear densities adjacent to the edge and at center of the lane.
Figure 7. Comparison of the air voids adjacent to the edge and at the center of the lane.
This section discusses the observations made during the paving and rolling operations that could have a significant impact on the performance of the Safety EdgeSM over time. As stated in the Introduction to the Field Report section, the objective of this field study was to evaluate the quality of the in-place HMA material and Safety EdgeSM by investigating three features.
During this site visit, field observations were limited to the paving of test sections #1 and #2 in which the contractor utilized the TransTech Shoulder Wedge Maker and the Carlson prototype #2. Neither device appeared to cause disturbances in the mat or with the shoulder material. Mix segregation in the finished overlay was not noticed in the test sections.
The contractor used a rubber tire Blaw-Knox PF-3200 paver and a Roadtec SB-2500C material transfer vehicle to overlay the existing milled HMA pavement. Test section #1 was paved first with the TransTech device mounted on the screed extension next to the end gate. Next, test section #2 was paved with the Carlson prototype #2 device which was quickly installed during a short break in paving. This device and Carlson's prototype #3 was a modified end gate with the angle of the Safety EdgeSM built into the end gate ski. The end gate ski was flat in the front and transitioned to 30° at the back of the ski. While paving with the Carlson prototype #2, the TransTech device remained mounted to the paver and was simply raised up and out of the way (Figure 8).
Figure 8. Paving with the Carlson prototype #2 device with the TransTech device is still attached but raised and out of the way.
Visual inspection of the slope faces of the first two sections revealed a coarse appearance with more gaps between exposed aggregate in section #1 than section #2 which had a smooth or sealed appearance. The smooth slope face produced by the Carlson device is thought to be a result of extruding the HMA over the length of the end gate ski into the Safety EdgeSM shape gradually from no slope near the front of the ski to 30° at the end of the ski.
Test section #3 was paved after the site visit and photos from WisDOT show a smooth slope face similar to section #2. An FHWA engineer on site indicated that before the Carlson Prototype #3 was sufficiently heated from the fresh HMA the slope face appeared only slightly smoother than the slope face produced by the TransTech device. It took roughly 200 ft of paving to heat up the end gate ski of the Carlson device after which the slope face became sealed and smooth. Future design modifications to the end gate design are expected to include a heating element to preheat and maintain the temperature of the device.
Regardless of which device was used, the shape of the slope faces were consistent throughout the test sections. One distinct difference among the edges was the slope face on section #2 produced by the Carlson Prototype #2 had a 0.25-inch vertical rise or lip as the slope face meets the horizontal surface of the mat. The lip may help retain the granular shoulder material. Figure 9, 10, and 11 show the three finished edges after compaction.
Figure 9. TransTech edge.
Figure 10. Carlson Prototype #2 edge.
Figure 11. Carlson Prototype #3 edge.
The contractor’s breakdown roller was an Ingersoll-Rand dual steel drum DD-110HF operating in low amplitude and high frequency mode (the roller vibrator control setting was set for a 2-inch mat). Typically, one vibratory pass was made hanging 12 inches over the free edge of the mat and 6 vibratory passes were made over the rest of the mat. No static passes were made by the breakdown roller. The intermediate roller was a Caterpillar PS-150B pneumatic tire roller that made 6 to 8 passes, none of which were at the edge. The finish roller was a Bomag BW 11 AS dual steel wheel roller. This roller was operated in static mode and made one pass hanging 6 inches over the free edge and 5 to 7 passes over the rest of the mat. The mat was stable and not overly tender under any of the rollers. No tearing or shoving was observed.
As stated above, the objective of this field study is to evaluate the quality of the in-place HMA material and Safety EdgeSM by investigating three features.
This section of the field report summarizes some of the findings and conclusions made during the paving/compaction operations.
The Safety EdgeSM should be inspected after the material for the shoulder has been placed to the final pavement elevation. In the long term, special attention should be focused on test section #2 to determine if the lip on this Safety EdgeSM is effective in retaining the shoulder material and if the smooth slope face on test section #2 and #3 impact the pavement performance.