Instrumentation Shows I-64 Segment II Pavement Recycling Project Meets Initial Expectations
In 2016, VDOT awarded the reconstruction and addition of new lanes along a portion of I-64 in Newport News, York, and James City Counties. The pavement design for this project, known as Segment II, included pavement recycling techniques, specifically full depth reclamation (FDR) of the existing pavement foundation for the reconstructed lanes and imported materials for the added lanes, and cold central plant recycling (CCPR) for the base asphalt layers.
VDOT based the design on lessons learned from previous pavement recycling projects, including the segment of I-81 in Augusta County constructed in 2011 and the VDOT-sponsored sections at the National Center for Asphalt Technology (NCAT) Test Track paved in 2012. The concept of using high-quality SMA surface layers, CCPR, and FDR has proven to provide long-lasting pavement sections in these high traffic applications. Figure 1 shows the cross-section for I-81, NCAT Section S12, and I-64 Segment II. A modification to the design used for I-64 Segment II is the inclusion of an asphalt-stabilized open-graded drainage layer (OGDL) to enhance drainage within the pavement structure.
To better understand the performance of the recycled pavement design used on I-64, a team of researchers from VTRC and VTTI installed instrumentation during the construction of the pavement section in 2018. The intent was to run trucks, loaded to known weights, over the instrumented section, and measure the pavement response. The instrumentation included pressure cells, strain gauges, thermocouples, and moisture sensors that were installed to quantify the pavement performance mechanistically. The instruments were placed in the right wheel path of the right lane in the westbound direction just past the ramp to Exit 242B (to northbound Marquis Center Parkway, SR 199). The contractor (Allan Myers) provided significant logistical and installation support to the researchers.
When instrumenting an asphalt pavement, the primary areas of focus usually include the mid-depth pavement temperature, the horizontal strain at the bottom of the asphalt layers, and the vertical pressure on the subgrade. The location of the instrumentation for I-64 Segment II is shown in Figure 2. For this pavement section, the strain sensors would ideally have been placed at the bottom of the OGDL layer. However, given that this layer was only 2 inches thick, it would not have been possible to place the 0.5-inch-thick strain gauges before the construction of the OGDL layer without causing problems for the creation of that layer. As an alternative, the researchers decided to place the strain gauges at the bottom of the CCPR layer and determine the maximum strain by modeling.
On five dates from June 2019 to December 2019, trucks from the VDOT Williamsburg Area Headquarters were loaded with sand or aggregate to the desired axle weights and were driven over the instrumented section by agency personnel as part of the public traffic stream. The researchers used a synchronized timestamp from a video camera, recording the passing traffic and the data acquisition system to determine when the test vehicle crossed the instrumented section, as shown in Figure 3.
Figure 4 shows the temperature normalized strain from those gauges located at the bottom of the CCPR layer. The data shown were normalized with respect to temperature because the mid-depth pavement temperature over the time period varied from approximately 92°F (June) to 51°F (December). Without this normalization, other performance trends would be masked by the change in stiffness with respect to temperature. The trend in Figure 4 shows that the strain decreases with respect to time. The researchers suggest that this decrease in strain could be due to continued curing of the recycled layers. Additional curing will increase the stiffness of the pavement section and thus reduce the measured strain. Previous work by VTRC showed that stiffening of a recycled pavement section by curing in the field could be observed at least one year from construction.
As noted previously, the maximum strain is expected at the bottom of the OGDL layer, below where the strain gauges were placed on Segment II. To calculate the strain on the bottom of the OGDL layer, the researchers modeled the pavement system using a layered elastic software program (WESLEA for Windows). The primary model inputs included the pavement layer thicknesses and the stiffness of each layer. The thickness values were obtained from cores that were collected after construction. Initial stiffness values were taken from previous research efforts by VTRC and adjusted until the modeled strain at the bottom of the CCPR layer matched the values measured from the field instrumentation. Figure 5 shows the strain distribution within the pavement structure based on the modeling results.
From Figure 5, it can be seen that the modeled strain at the bottom of the OGDL layer is less than approximately 50 microstrains, well within limits shown to result in a very long-lasting pavement. This compares favorably with the temperature normalized strain values from Section S12 at NCAT shown in Figure 6. Given that the pavement structure for Segment II is thicker than Section S12, a lower strain value is expected. Using these initial measurements, the researchers can begin to monitor the I-64 recycled pavement and determine the relative health and expected life of the pavement section.