Pavement Testing Facility Overview
The Pavement Testing Facility (PTF) uses rapid pavement testing of full-scale structures to evaluate the durability of both new and existing pavement materials. Recent testing has contributed to the development and verification of new specifications, designs, and test procedures for rigid and flexible pavements. Past and potential future applications include assessing the impact of different tire or load configurations on pavement performance.
The facility simulates truck traffic with controlled loading and pavement temperatures. In only a few months, the facility’s loading machines can apply wheel loads comparable to many years of service, and collect the corresponding pavement distress and performance data. Two machines allow simultaneous testing of two pavement lanes under the same ambient temperature and moisture conditions or at the same pavement age.
The PTF Lab’s main capability is full-scale testing of pavement structures. There are 12 testing lanes. Each full-scale test lane is 50 by 4 meters (165 by 13 feet), and can be further divided into 4 subsites, for a total of 48 test locations on the facility grounds.
A supplemental three cell test pit is available for material testing. The test pit contains a reaction frame for load plate testing. All cells are enclosed; this allows the water table to be controlled with additional flexibility in testing unbound materials for pavement structures and other geotechnical applications.
Services are directed to research and investigations at Turner-Fairbank Highway Research Center (TFHRC) or in cooperation with other government laboratories, agencies, research institutions or associations, with the objective of advancing the knowledge of pavement engineering and the performance of highway materials and structures.
Current Full-Scale Projects
SUSTAINABILITY: Reclaimed Asphalt Pavement
Advance Use of Recycled Asphalt in Flexible Pavement Infrastructure: Develop and Deploy Framework for Proper Use and Evaluation of Recycled Asphalt in Asphalt Mixtures
Objective: The key project objective is to quantify the cracking resistance of high recycled asphalt pavement (RAP) mixtures that considers the use of lower temperature production with warm-mix asphalt (WMA) and recommend any limitations for combining the two technologies.
Approach: Construct and load 10 test pavements with different quantities of RAP, different WMA technologies, and different production temperatures at the Pavement Testing Facility. The relative ranking in fatigue cracking as well as supporting laboratory characterization will guide and generate recommendations.
Status: Eleven sites were constructed with ten different mixtures. Eight sites have been successfully tested. Two remain to be completed during spring 2017.
The current full-scale experiment was built by FHWA in 2013. It contains 10 different test lanes with varying amounts of recycled material content. The objective is to evaluate the fatigue cracking performance of sustainable asphalt materials and mix designs to establish realistic boundaries for high content RAP and reclaimed asphalt shingle (RAS) mixtures employing WMA technologies based on percent binder replacement and binder grade changes. All lanes consist of 100 mm (4 inches) of asphalt concrete (various mixtures), over 560 mm (22 inches) of unbound aggregate base, over the existing subgrade. A line of geotextile separates the base from the subgrade. The base is uniform across the entire paved testing area. The only variable in the current experiment is the surface asphalt concrete mixture.
The mixture experimental design, provided in table 1, includes two virgin asphalt binders (PG 64-22 and PG 58-28), four recycled binder ratio (RBR) levels (0, 20, and 40 percent from RAP and 20 percent from RAS), and two production temperatures (300–320 °F for hot mix asphalt (HMA) and 240–270 °F for WMA). The two WMA technologies were water foaming and a chemical binder additive. A PG64-22 binder is the default performance grade (PG) for the mid-Atlantic location of the project and a PG58-28 virgin binder was chosen to offset the addition of RAS and high-RAP. The asphalt layer for each ALF lane was built in two lifts that were nominally 2 inches thick and compaction was monitored with a nuclear density gauge.
Before asphalt paving in the ALF test lanes, the granular aggregate base (GAB) had to be reconditioned after the previous experiment’s asphalt layer was milled and removed. Approximately the top 8 inches of GAB were scarified with a motor grader, supplemented with water when necessary to achieve the optimum moisture content, and compacted with a smooth drum vibratory roller. The compaction was controlled using a nuclear density gauge tested in at least six different locations in each ALF Lane to meet or exceed 95 percent of the modified Proctor dry density. Based on historical trenching, the subgrade was presumably not disturbed by past experiments, and therefore not reconditioned.
MATERIALS: Asphalt Concrete Field Density and Unbound Material Geosynthetic Reinforcement
Investigation of Asphalt Concrete Compaction and Its Impact on Performance of Pavements Built with and without Geosynthetic Base Reinforcement
Objective: Construction variability is an intrinsic characteristic of pavement construction. The compaction of asphalt concrete (AC) mixtures is a critical component in the process of achieving optimal performance. The objective of this research is to look into various levels of AC compaction and how it affects pavement performance. Cracking and rutting field experiments will be conducted in three lanes built at three different AC density levels. In addition, half of the lane area will be constructed with a geosynthetic base reinforcement. This additional variable will provide insights into the pavement performance when the unbound base layer is subjected to a geosynthetic reinforcement.
Approach: Construct three lanes with one AC mixture at three different densities. One half of each lane will have a geosynthetic membrane installed at the midpoint of the base layer. The other half will not. Each lane will be loaded at the accelerated loading facility (ALF). Cracking and rutting will be monitored in two separate temperature controlled loading tests.
Status: Construction completed at the beginning of fall 2016. Loading will start in spring 2017.
The current full-scale experiment was built by FHWA in 2016. It contains four different test lanes with the same pavement structure and materials, varying only the compaction level of the asphalt concrete layer. The purpose is to evaluate the impact on rutting and cracking performance of different levels of compaction obtained during construction (field density). All lanes consist of 100 mm (4 inches) of asphalt concrete (same mixture), over 560 mm (22 inches) of unbound aggregate base, over the existing subgrade. A geosynthetic base reinforcement was installed 150 mm (6 inches) from the top of the base layer, within the crushed aggregate base. The experimental design includes two independent variables: the field density of the asphalt concrete layer, and the presence or not of the geosynthetic reinforcement.
Before asphalt paving in the ALF test lanes, the granular aggregate base (GAB) had to be reconditioned after the previous experiment’s asphalt layer was milled and removed. Approximately the top 12 inches of GAB were removed. The remaining material was scarified with a motor grader, supplemented with water when necessary to achieve the optimum moisture content, and compacted with a smooth drum vibratory roller. Six inches of new crushed aggregate base was placed and compacted. The geosynthetic base reinforcement was installed in half of the area of each lane. The base layer was completed with the placement and compaction of new crushed aggregate material on the last 6 inches. The compaction was controlled using a nuclear density gauge tested in at least six different locations in each ALF Lane to meet or exceed 95 percent of the modified Proctor dry density. Finally the asphalt concrete was laid and compacted to various density levels.
- Twelve full-scale pavement test lanes.
- Two Accelerated Loading Facility (ALF) machines.
- Falling weight deflectometer (FWD).
- Portable seismic pavement analyzer (PSPA) used for nondestructive pavement testing.
- Light-Weight Deflectometer (LWD)
- A semiautomatic laser surface profiler to measure both transverse and longitudinal pavement surface profiles.
- Dynamic cone penetrometer.
- Instruments, sensors and equipment that:
- Measures load-associated pavement response (stress, strain, and deformation), and environmental effects (temperature and moisture).
- Include a state-of-the-art multichannel data acquisition system to collect pavement instrumentation response data and pavement temperature.
- Can record real-time temperature data for temperature-controlled loading.
- A layer deformation measuring system to monitor vertical compressive strain and rutting in each layer of pavement.
- Computer workstations and software to perform advanced pavement analysis and material modeling, as well as mechanistic design.
- Relational databases developed to provide customers with a variety of data from pavement-testing experiments, especially pavement response and performance data.
The ALF is a transportable linear full-scale accelerated loading facility which applies a rolling wheel load on a 13.7 m (45 ft) test length of any test pavement. Figure 1 shows two ALF machines at the Federal Highway Administration (FHWA) PTF. The first ALF was delivered to PTF in 1986 and the second was purchased in 1993.
Figure 1. FHWA’s two ALF machines loading pavements at PTF site.
The ALFs simulate traffic at controlled loading and pavement temperature conditions. An infrared heating system and thermocouples in the pavements provide the required pavement temperature. The ALF frame is 29 meters long (95 ft) and about lane-width (3.6 m or 12 ft). Each machine is capable of applying an average of 35,000 wheel passes per week from a half-axle load ranging from 33 kN (7,500 lbf) to 84 kN (19,000 lbf). An electric motor provides unidirectional loading passes at constant speed programmable up to 18 km/h (11 mi/h).
Radiant heaters are used to maintain an ambient temperature control; temperatures as high at 74 °C (165 °F) can be generated to accelerate asphalt rutting and during the fall, spring, and winter.
Currently, both machines are equipped with super single (425/65R22.5 wide base) tires. Previous research illustrated that a wide base tire type 425 induces greater damage than conventional dual tires. This provides a time savings advantage in accelerated loading. In addition, the simplicity of a single wheel has some advantages in mechanistic pavement analyses. A typical tire footprint is shown in figure 2, for a wheel load of 71 kN (16,000 lbf) and tire pressure of 827 kPa (120 lbf/in2). Lateral wheel wander of the ALF transverse position is programmable. Three standard deviation tables are available: zero wander, 50 mm (1.74 inches) and 133 mm (5.25 inches) standard deviation. Figure 3 illustrates the loading carriage from the pavement point of view.
Figure 2. Diagram of 425 tire imprint. Scaled outline of six tire treads
imprinted as vertical bars with circles overlaid on top illustrating the
relative size of an a calculated uniformly loaded circular area
and effective size of a circular contact based on the imprint area.
Figure 3. ALF loading carriage from the pavement point of view.
Pavement Test Sites and Instrumentation
There are 12 pavement test lanes in the facility, each 4 m (13 ft) wide. The paved area is 48 by 50 m (156 by 165 ft). Each lane contains four testing areas, designated as sites, shown in blue in figure 4. Each site has an effective loading area of 13.7 by 1 m (45 ft by 40 inches).
Figure 4. General overview of the pavement testing sites.
The effective loading area includes tire wander (if used in the experiment) and excludes the transitioning zone used to lower and raise the tire. The effective loading area is also the area in which cracking and all performance measures are taken during the loading phase of the experiment. In each lane there are three areas designated for sampling materials for laboratory testing. The detailed view of one typical lane is provided in figure 5.
Figure 5. Typical lane layout.
Pavement performance data is collected routinely at a targeted number of load passes defined by the experiment plan. Usually these include transverse profile, cracking, and rutting as distress measurements. In addition, structural nondestructive testing, using the LWD and PSPA, is performed to evaluate in situ stiffness.
All measurements are recorded in a database. Cracking is registered manually by drawing crack maps. These maps are later digitized for analysis and documentation. Rutting is measured through the laser surface profiler (total rutting only) and also through the rod and lever survey. Aluminum survey plates are installed on the top of aggregate base during the construction. Reference rods are attached to the aluminum plates before testing to measure the permanent deformation on top of base during the ALF loading. Figure 6 shows this layer deformation measurement assembly. The elevation of the top of each reference rod will be recorded with a rod and level. Multiple-Depth Deflectometers (MDDs) can also be installed for multiple layer deformation measurements. Thermocouples are normally installed just outside the effective testing area. They provide temperature readings that trigger the controllers of the heaters to keep the surface layer temperature (usually mid-thickness) at a constant prescribed value. This is particularly important for asphalt concrete testing. Figure 7 illustrates a typical layout for performance measurements.
Figure 6. Layer Deformation Measurement Assembly.
Figure 7. Survey Plate and Thermocouple Locations for a typical test site (not to scale).
In addition to cracking and deformation/displacement measurements, embedded gages can be installed for monitoring strain and stress developments within the pavement structure. These devices can be installed anywhere in the structure and are connected to a data acquisition system for periodic monitoring.
Falling Weight Deflectometer
The Falling Weight Deflectometer (FWD) is capable of generating a transient, impulse-type load of 25–30 milliseconds in duration, at any desired (peak) load level between 680 and 12,250 kgf (1,500 and 27,000 lbf). Figure 8 shows the FWD system during testing at the ALF. The deflections induced by the dynamic load are determined using geophone data that measure the velocity of the load pulse as it travels through the layered media. The velocity signal is integrated to determine the deflections induced during loading at various locations. These deflections are used in association with the applied load to calculate in situ stiffness of the layered structure. This process is called backcalculation.
Figure 8. FWD Test system operating on top of reconditioned ALF aggregate base.
Figure 9. Closeup of the loading plate on the FWD test system.
Figure 10. Closeup of the surface geophones on the FWD test system.
Portable Seismic Pavement Analyzer (PSPA)
The PSPA uses wave propagation techniques to measure fundamental material properties. Waves induced by vibrations are measured by sensors and used to determine the modulus of the top layer. The PSPA can also be used to determine the thickness of the top layer under certain conditions, and defects such as voids, cracks, and zones of deterioration through an impact echo test. The PSPA operates with a laptop computer connected via cable to the hand-carried transducer unit. Two accelerometers and a high frequency source are deployed in the sensor unit, as shown in figure 11. The collection and preliminary reduction of data at one point take less than 15 seconds. Material characterization by PSPA can be conducted on bound pavement layers (like asphalt and concrete) and soil-type materials (e.g., aggregates).
Figure 11. Portable Seismic Pavement Analyzer.
Light Weight Deflectometer (LWD)
The Light Weight Deflectometer (LWD) is a portable version of the FWD. The LWD can be used to test thin asphaltic pavements, recycled materials bound with foamed bitumen, and directly test the unbound subbase and subgrade. The output from the LWD can be used to calculate the strength of multiple pavement layers and perform quality control/quality assurance (QC/QA) tests for compaction quality.
Figure 12. Light Weight Deflectometer.
Dynamic Cone Penetrometer (DCP)
The Dynamic Cone Penetrometer (DCP) is an instrument that can be used for the rapid measurement of the in situ strength of existing flexible pavements constructed with unbound materials. Measurements can be made down to a depth of 800 mm (31.5 inches) or to a maximum depth of 1500 mm (59 inches) by adding an extension rod. Moreover, where the pavement layers have different strengths, the boundaries between them can be identified and the thickness of each layer determined through analytical or manual data interpretation. The DCP uses an 8 kg hammer dropping through a height of 575 mm (22.6 inches), which drives a 20-milimeter (3/4-inch) rod with a 60 degree cone tip into the pavement layer. Figure 13 shows the DCP and its main parts.
Figure 13. Dynamic Cone Penetrometer.
Laser Surface Profiler
The laser surface profiler is used to measure the surface transverse profile. The laser device is mounted in a trolley that runs on top of a reference system, as shown in figure 14. This facilitates computing the distance from the reference system to the top of the pavement surface. The measurements are performed in semi-automated mode; the trolley is pulled manually and the distances are recorded automatically. The transverse profile indicates the surface deformation, used to monitor permanent deformation, as illustrated in figure 15. This device can also be used to measure the longitudinal profile to determine the overall roughness of the tested pavement.
Figure 14. Laser Surface Profile and with closeups.
Figure 15. Example of a transverse surface profile after 100,000 load passes.
Weather Station Unit
The PTF Laboratory has a local weather station. It is a small unit, as seen in figure 16. The station is solar energy powered and capable of measuring several weather related data.
Figure 16. PTF weather station unit.
A variety of instruments and sensors are available to measure load-associated pavement response (stress, strain, and deformation). Probes and thermocouples measure environmental conditions (temperature and moisture), including a multichannel data acquisition system to collect pavement instrumentation response data, electronic temperature control, and data acquisition to show and record pavement temperatures at various locations in real time.
The facility has several computer workstations and software programs to perform advanced pavement analysis and material modeling, as well as mechanistic design, and relational databases developed to provide customers with a variety of data from pavement testing experiments, especially pavement response and performance data.
- Li, X., and Gibson, N. (2016). Comparison of Laboratory Fatigue Characteristics with Full-Scale Pavement Cracking for Recycled and Warm-Mix Asphalts. Transportation Research Record: Journal of the Transportation Research Board, Vol. 2576, pp. 100–108.
- Mbarki, R., Kutay, M. E., Gibson, N., and Abbas, A. R. (2012). Comparison between fatigue performance of horizontal cores from different asphalt pavement depths and laboratory specimens. Road Materials and Pavement Design 13, No. 3, pp. 422–432.
- Li, X., Gibson, N., Qi, X., Clark, T., and McGhee, K. (2012). Laboratory and Full-Scale Evaluation of 4.75-mm Nominal Maximum Aggregate Size Superpave Overlay. Transportation Research Record: Journal of the Transportation Research Board, Vol. 2293, pp. 29–38.
- Gibson, N., Qi, X., Shenoy, A., Al-Khateeb, G., Kutay, M., Andriescu, A., Stuart, K., Youtcheff, J. and Harman, T. (2011). Full-Scale Accelerated Performance Testing for Superpave and Structural Validation: Transportation Pooled Fund Study TPF-5 (019) and SPR-2 (174) Accelerated Pavement Testing of Crumb Rubber Modified Asphalt Pavements. Federal Highway Administration, Report No. FHWA-HRT-11-045, McLean, VA.
- Kutay, M.E., Gibson, N.H, and Youtcheff, J. (2008). Conventional and Viscoelastic Continuum Damage (VECD)-Based Fatigue Analysis of Polymer Modified Asphalt Pavements. Journal of the Association of Asphalt Paving Technologists (AAPT), Vol. 77, pp. 395–434.
- Kim, Y. R., Guddati, M. N., Underwood, B. S., Yun, T. Y., Subramanian, V., and Savadatti, S. (2009). Development of a multiaxial viscoelastoplastic continuum damage model for asphalt mixtures. Federal Highway Administration. Report No. FHWA-HRT-08-073, McLean, VA.
- Al-Khateeb, G., N. Gibson, and X. Qi (2007). Mechanistic analyses of FHWA's accelerated loading facility pavements: Primary response. Transportation Research Record: Journal of the Transportation Research Board, Vol. 1990, pp. 150–161.
- Qi, X., Mitchell, T., Gibson, N., and Harman, T. (2006). Pavement Responses from the Full-Scale Accelerated Performance Testing for Superpave and Structural Validation. Pavement Mechanics and Performance, pp. 75–86.
- Stuart, K. D., Mogawer, W. S., & Romero, P. (2002). Validation of the superpave asphalt binder fatigue cracking parameter using an Accelerated Loading Facility. Federal Highway Administration. Report No. FHWA-RD-01-093, McLean, VA.