Printer Friendly

The FHWA test road: construction and instrumentation.


Three valuable resources give the Federal Highway Administration (FHWA) the capability to perform full-scale local tests on highway pavements. These resources are located at the Turner-Fairbank Highway Research Center (TFHRC) in McLean, Virginia:

* The Pavement Test Facility, which includes the fixed-speed rolling wheel Accelerated Loading Facility (ALF).(1,2) [1] This facility provides information on ultimate response (e.g., fatigue cracking and rutting) of flexible pavements in a matter of months; thus, problems and likely modes of failure can be identified long before they would occur on inservice highways.

* The Pavement Isothermal Test System (PITS), which consists of two concrete chambers containing full-scale pavements. These pavements can be subjected to any configuration of controlled stress fixed-position loading in order to simulate a variety of highway load magnitudes and frequencies.

* The Test Road Facility, which is used to monitor the pavement's primary response when it is subjected to different types of moving axle suspension systems and truck configurations.

The FHWA is now constructing a fourth pavement test facility, a prototype dynamic truck actuation laboratory (DYNTRAC) to study the nature of dynamic forces on pavements induced by heavy trucks. DYNTRAC will measure the forces exerted by truck wheels as they are subjected to vertical oscillations simulating known road profiles.

Together, these facilities (see figure 1 ) will provide the basis for pavement research throughout the nineties. Studies at these laboratories, coordinated through the high priority research areas of Accelerated Evaluation of Pavement Performance and Truck-Pavement Interaction, will provide solutions to pavement infrastructure deterioration.

This article discusses the Test Road Facility constructed during the summer of 1990. The first test series, "Test Series I" was conducted in August of 1990. Data gathered from these tests were analyzed. Results to date are presented in three separate documents in terms of primary response load equivalency factors, primary deflection responses, and primary layer horizontal strain responses. (3,4.5)

Test Road Facility Overview

The Test Road Facility contains two full-scale flexible pavement test sections capable of accommodating any variety of vehicle types traveling at speeds up to 88 km/h (55 mi/h). The main feature of the Test Road is strain and deflection instrumentation-strategically located within the pavement sections--that measures primary responses to moving truck loads,

Experiments performed on the Test Road will study the effects of tire pressure and type, axle configuration and weight, suspension system type, truck configuration, and other truck traffic factors on pavement performance. Different combinations of these factors can cause greater or lesser wear to the highway pavement.

The data from the Test Road will be used to support the Truck-Pavement Interaction high priority area goals. These include:

* Verify layer theory and finite element primary response models.

* Develop primary response load equivalency factors, in terms of the standard 80.100 N (18kip) axle, that express pavement damage attributed to vehicles with different weights and load distributions.

* Determine dynamic effects experienced by pavements as a result of the impact loading generated by the bouncing motion of vehicles induced by rough pavements.

Site Selection

Factors considered in choosing the Test Road site included safety, alignment, traffic volume, environmental impacts (such as noise), office and laboratory accessibility, and proximity to existing utilities. The site shown in figure 2 was selected because of its relatively straight alignment which permitted safer operation of trucks at maximum test speed,

A site investigation, which primarily entailed core drilling, was conducted to determine the in situ pavement conditions and estimate excavation quantities. Asphalt pavement thickness was found to be 165 mm (6.5 in); the base material was 381 mm (15 in) of bank-run gravel. The subgrade was a silty sand, classified as A-4, with a California bearing ratio (CBR) value of 5.


The Test Road design only required reconstruction of the inbound lane of the access road connecting Virginia Route 193 with the TFHRC. The Test Road was designed to contain two 30-m (100ft) test sections separated by a 7.6-m (25-ft) transition zone shown in figure 3. The thick pavement test section consisted of an A-4 subgrade overlaid with 152 mm (6 in) of A-4 subbase, a 305-mm (12in) crushed stone base course, and topped with 178 mm (7 in) of asphalt concrete pavement. The thin section consisted of the same A-4 subgrade and 152 mm (6 in) of A-4 subbase overlaid by 305 mm (12 in)of crushed stone and a 89 mm (3.5-in) asphalt concrete pavement. A crushed stone shoulder3.5 m (11 ft) wide and up to 0.6 m (2 ft) thick was added along the Test Road to serve as a working platform. Figures 3 and 4 illustrate the longitudinal and transverse cross sections of the road respectively. It is important to note that the Test Road thickness design is identical to the ALF Phase II thickness design. Construction materials included recycled asphalt concrete mix and dense graded crusher run limestone. The recycled asphalt concrete was a blend comprising about 85 percent virgin material. Lime was added to the asphalt concrete mix in accordance with Virginia Department of Transportation specifications, to reduce stripping (6).

The test sections were designed to accommodate strain and deflection instruments placed along the center of the left wheel path of each section; the transition zone was designed as a utility junction for the wires.


To construct the Test Road the old asphalt surface and bank-run gravel base had to be removed, and an additional 152 mm (6 in) of subgrade excavated, to eliminate any possible contamination of the subgrade and base course. This work entailed sawcutting 76.2 m (250 ft) of pavement and removing 152.9 [m.sup.3] (200 [yd.sup.3]) of old material in preparation for placing the new pavement. Additional A-4 material was used to replace the mixed soil and bank-run base. This material was then rolled to seal it against water intrusion when placing the soil moisture gauges.

The crushed stone base was placed in 152-mm (6in) lifts to a depth of 305 mm (12 in) and compacted to 2 290.9 kg/[m.sup.3] (143 pcf). Construction then stopped for about a week as the strain gauges and parts of the single-point deflectometers were installed.

Next, a trenching machine was used to cut trenches to a instrumentation depth of about 0.6 m (2 ft) along the shoulder-pavement interface; these trenches were for collecting the wires in groups from feeder trenches.

One of the challenges of installing the instruments was ensuring that they were not damaged by the heat or vibration generated by paving and that the leads were buried to protect them from damage. Thirty-three H type strain gauges had their leads buried about 0.05 m (2 in)into the crushed stone base. This design was considered safe, but was labor-intensive and could potentially disturb the integrity of the compacted stone base. In an effort to find an easier way to install the instrument leads without disturbing the previously compacted stone-base material, the leads of three other H-type gauges were wrapped in silicon-impregnated fiberglass tubing to prevent punctures from stones and to insulate the leads from the heat. The leads were then placed on top of the compacted base material where, for additional protection, they were covered with hand-placed hot mix asphalt about 1 hour before paving. Since these gauges functioned normally after paving, this method seemed to be effective. While the instruments were being placed, utility work was undertaken to provide electricity to the site.

Paving was done with a rubber-tired paver capable of straddling the instruments. The sawcut edges of the existing pavement were tack coated with an emulsion, as was the priming of the crushed aggregate base, to form a seal and bond the new surface to the existing pavement. Surface mix material was hand placed to cover some of the gauges. Paving progressed from the thick to the thin section, and was placed in 76-, 51-, or 38-mm (3-, 2-, or 1.5-in) lifts as appropriate. The paver's vibrating screed was used to develop a compaction density of 2 579.2 kg/[m.sup.3] (168 pcf). The construction did not damage the instrumentation that had been installed.


The Test Road's experimental sections were designed to collect primary response measurements (deflection, strain) and environmental data (moisture, temperature). These parameters are typically needed to verify the use of mechanistic models in pavement design and life-cycle analysis. The baseline pavement profile was measured at completion of each layer to determine the variability of construction operation and provide initial elevation measurements for layer theory analysis. Table 1 shows the gauges installed, the parameters they measure, and the experiments in which they are used.

Pavement response is measured with strain gauges and deflectometers. The gauges installed were either the H-type (Kyowa) or a variety made by the Alberta Research Council (ARC). The deflectometers consisted of both single-layer and multilayer types and were installed in each section. The locations of the various instruments are shown in figure 5. A graphic display of each gauge is given in figure 6 a through f.

The H-type gauges were installed to measure the dynamic loading of the thick pavement section. To ensure their adherence to the underside of the asphalt pavement while retaining alignment along the path, the gauges were modified at the TFHRC Highway Electronics Laboratory. They were installed in a line of 36 gauges set 0.305 m (1 ft) apart.

The ARC strain gauges were placed to study strain in the pavement under controlled loading conditions. These gauges were placed in a pattern to cover the vehicle wheel print. A configuration of two rows of two in each section was used for redundancy. Deflectometers were installed to evaluate the degree to which the pavement layers deflect. The single-layer or single-point deflectometers, also supplied by ARC, were used in coordination with the ARC strain gauges. Multidepth deflectometers were used to evaluate the compressive strains occurring in the lower structural layers. These instruments were fabricated at Texas A&M University. Both types of deflectometers are linear variable differential transformers; they were anchored at a depth of about 3 m (10 ft) to isolate them from vibration, noise, and settlement.

The environmental gauges were used to correlate the weather and climatic conditions to the responses measured. Soil moisture was measured using gauges placed 152 mm (6 in) and 305 mm (12 in) into the subgrade under each section.

Thermocouples were installed to measure the temperature changes in the pavement during testing. Each section contained one set of instruments; these provide temperature information at seven locations from the pavement surface to a depth of 1.3 m (4 ft).

Truck-Pavement Interaction

The Test Road was developed to support the overall goals of the FHWA's high priority area on Truck-Pavement Interaction. The three phases are summarized:

* Load Equivalency Factors (LEF's).

* Primary Response Analysis.

* Vehicle/Pavement Interaction.

Accomplished work in each phase will depend on the results from test conducted on the Test Road, ALF, PITS, and DYNTRAC.

Phase I: Load Equivalency Factors

This phase is substantially complete. Work comprised the design and construction of the Test Road, the conduct of Test Series I, and an evaluation report on the concept of primary response load equivalency. (3)

Initial testing was conducted in August 1990 on the Test Road's two instrumented test sections. The factors and levels included in the experiment are shown in table 2. The three vehicle types and axle loads used in the testing are shown in table 3. The single unit two-axle vehicle was used to apply the standard 8 172-kg (18-kip)load. The measured pavement primary responses were used, along with several selected primary response LEF methods, to develop LEF's and evaluate the accuracy and reliability of alternative methods (3). Some significant conclusions were:

1. Load level was the most significant factor.

2. Pavement thickness was relatively insignificant.

3. Slower vehicle speeds showed higher damage potential.

4. At the low and medium levels of axle loads, tire pressure had very little effect.

5. Deflection-based LEF methods are reasonable.

Phase II: Primary Response Analysis

This phase of the research deals with the effects of load characteristics on the primary response of the pavement system and with the ability to estimate this occurrence using elastic and viscoelastic theory.

To determine the influence of vehicle parameters (load axle type, tire pressure) on pavement primary response, portions of the experimental data from Test Series I and II will be used. These responses will be analyzed in terms of the time-dependent and -independent plastic and viscoelastic properties of the pavement materials.

Phase III: Vehicle/Pavement Interaction

This effort will develop results from Test Series II at the Test Road, full-scale field tests at the ALF, laboratory tests using DYNTRAC, and computer simulations.

In Test Series II and field tests several types of trucks with different suspensions will be run on the Test Road. The trucks will be equipped with wheel force transducers. Bumps will be placed on the pavement so that response data can be obtained from the 36 imbedded H strain gauges.

Results from tests conducted on the DYNTRAC system will emulate the FHWA Test Road profile; these will be compared with the responses from the tests conducted on the Test Road. Resuits from the ALF on dual versus single tires will be combined with predictors of flexible pavement testers to validate a new mechanistic submodel of VESYS.


Predicting pavement responses under surface loads is an important component of the mechanistic analysis of flexible pavements. These responses are a function of both the pavement structure and load characteristics. Factors such as tire pressure, axle load, axle configuration, vehicle speed, layer thickness, and time-dependent material properties influence the response. The TFHRC Test Road was constructed to evaluate these functions as they relate to pavement design. The experiments conducted at this facility promise to provide important insight that will improve existing knowledge of the interactive relationship between vehicle performance and pavement behavior.


(1) David A. Anderson, Walter P. Kilareski, and Zahur Siddiqui. Pavement Testing Facility: Design and Construction, Publication No. FHWA-RD-88-059, Federal Highway Administration, Washington, DC, August 1988.

(2) David A. Anderson, Peter Sebaaly, Nader Tabatabaee, Ramon Bonaquist, and Charles Churilla. Pavement Testing Facility: Pavement Performance of the Initial Two Test Sections, Publication No. FHWARD-88-060, Federal Highway Administration, Washington, DC, December 1988.

(3) Stuart W. Hudson, Virgil L. Anderson, Paul Irick, and R. Frank Carmichael III. Impact of Truck Characteristics on Pavement Truck Load Equivalency Factors, Publication No. FHWA-RD-91-064, Federal Highway Administration, Washington, DC, March 1991.

(4) W.J. Kenis and Gustav Rhode. "Primary Response Under Heavy Truck Traffic," paper presented at the Seventh International Conference on the Design of Asphalt Pavements, Cambridge, England, August 1992.

(5) W.J. Kenis. "Flexible Pavement Strain Response Under Moving Truck Traffic." paper presented at the Engineering Foundation Conference: Vehicle Road Interaction II, Santa Barbara, CA, June 1992.

(6) Road and Bridge Specification, Virginia Department of Transportation, Richmond, VA, January 1987.

Kevin Black is a highway engineer in the Materials Branch, Construction and Maintenance Division, of the Federal Highway Administration (FHWA). Formerly, he was with the Pavements Division in the Office of Research and Development.

William Kenis is Program Manager for FHWA's Truck/Pavement Interaction high priority area in the Pavements Division, Office of Research and Development.
 Table 1.--Summary of instruments and functions
Instrument Measured property Experiment
"H" strain gauge Strain Dynamics
ARC strain gauge Strain Response
Single layer Deflection Response
Multilayer Deflection Response
Moisture cell Moisture content
Thermocouple Temperature
 Table 2.--Factors and levels in experiment
Factors Levels
Pavement structure--(PVMT) weak, strong
Instruments nested in 4-sets (a duplicate set
pavement--(INST) in each pavement
Axle type Single, tandem,
Axle load Low, Medium, High
Tire pressure 515-kPa (75-psi)
 760-kPa (110-psi)
Speed 8-km/h (5-mi/h)
 72-km/h (45-mi/h)

COPYRIGHT 1992 Superintendent of Documents
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1992 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Federal Highway Administration
Author:Black, Kevin; Kenis, William
Publication:Public Roads
Date:Jun 1, 1992
Previous Article:Rock riprap for protection of bridge abutments located at the flood plain.
Next Article:An investigation of passing accidents on two-lane, two-way roads.

Related Articles
Technology applications.
Test roads: designing the pavements of the future.
Utah Receives $26 Million for 2002 Winter Olympics.
Open road for crushers. (C&D News).
Albert F. DiMillio selected to receive prestigious Engineering Award. (Personnel).
Editor's notes.
Robust partnerships create a stronger transportation system. (Editor's Notes).
FHWA cites top traveler information Web sites. (Management and Administration).
New course explores geotechnical aspects of pavement design.
FHWA receives Recycling Award.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters