Back to Basics

Nondestructive Testing of Pavement Structures

by Nishantha Bandara* and Robert C. Briggs*

 

Determining the condition of pavement structures is not easy. It is also extremely important to our daily lives and helps determine the amount we spend on pavements and their repair. Here is an introduction to the basics of the subject - an easy to read, valuable "Back to Basics." Frank Iddings Tutorial Projects Editor

 

INTRODUCTION

The United States' transportation network incorporates over 15 000 airports and more than 6 44 0000 km (4 000 000 mi) of highway pavements. This pavement network forms a significant portion of the national infrastructure and represents a cumulative investment of billions of dollars over several decades. Hence, there is a growing need for strategic management of the national roadway and airport pavement network to preserve this large capital investment. Typically, pavement management activities require accurate, fast and cost effective pavement test techniques to record current status of the pavement structure. NDT techniques provide powerful tools to test pavement structures in a rapid manner. Significant advances have been made in the last few years on NDT of pavement structures. These techniques include nondestructive deflection testing, ground penetrating radar and spectral analysis of surface waves. The purpose of this paper is to present commonly used technologies in the NDT arena and their application in pavement testing. Data analysis for ground penetrating radar and falling weight deflectometers are presented in technical papers in this issue, however.

 

NONDESTRUCTIVE DEFLECTION TESTING
Nondestructive deflection equipment operates by applying a load to the pavement surface and measuring the resulting surface deflections using velocity transducers or geophones. The results obtained from NDT equipment are used for determining the following in asphalt and concrete pavements:

• asphalt pavements - elastic modulus of pavement layers; pavement structural adequacy; overlay thickness design; load limits; remaining structural life
• concrete pavements - concrete elastic modulus and subgrade modulus of reaction; load transfer across joints; void detection under concrete slabs; pavement structural adequacy; rehabilitation design.

At present there are many types of deflection testing equipment and they can be categorized into three basic types: static, steady state dynamic and impulse. Each of these types of deflection equipment are described below.

---

Nondestructive testing of pavement structures has gained popularity in the recent past.

---

 Static or Slow Moving Deflection Equipment
Static deflection equipment is used to measure pavement surface deflections under static or slow moving loads. The most commonly used static equipment includes the benkelman beam. This provides deflection measurements at any number of points under a nonmoving or slow moving load (similar to an "influence line" used in structural testing of buildings and bridges). This device was developed at the Western Association of State Highway Organizations Road Test in 1952 and was the most widely used device until recently. The benkelman beam test procedure involves the measurement of a pavement surface rebound with a cantilevered beam as a truck loaded to 80 kN (18 000 lb) on its rear axle moved from rest. Measurements are made between dual tires on the rear axle at specified intervals in the outer wheel path.

Several versions of automated benkelman beams were developed by various manufacturers, typically by mounting the deflection beams on the truck that provides the axle load. The truck moves slowly, in the range of 1.6 to 3.2 km/h (1 to 2 mi/h). The main advantages of these static/slow moving deflection testing devices are simplicity, low instrument cost and the possibility of utilizing realistic load levels. The disadvantages of these devices are that they are slow, labor intensive, do not provide a "true" deflection basin and suffer relatively poor precision and bias.

Steady State or Dynamic Vibratory Equipment
Steady state or dynamic vibratory equipment uses a relatively large static preload and a sinusoidal vibration to the pavement with a dynamic force generator. With some devices, it is possible to change the magnitude and the frequency of the applied load. A major problem with this equipment is that the relatively large static preload may adversely affect the accuracy of the test.

One of the first commercially available steady state deflection measuring devices is trailer mounted and can be transported with any standard automobile. Testing is standardized by AASHTO T256-77 and ASTM D4695-87. The system must be stationary when measurements are taken. The force generator (counter rotating weights) must be started subsequent to lowering the deflection sensors (velocity transducers) to the pavement surface. The maximum peak to peak dynamic force is 454 kg (1000 lb) at a fixed frequency of 8 Hz. The load is applied through two 102 mm (4 in.) wide, 406 mm (16 in.) diameter rubber coated steel wheels which are placed 508 mm (20 in.) apart. The system is highly reliable (low maintenance) and can produce a full deflection basin. Disadvantages include a significantly low dynamic load, relatively large static preload, susceptibility to errors due to pavement resonance effects and inadequate dynamic load to test heavy pavements. Standard loading and deflection measurement locations are shown in Figure 1.

Figure 1  - Standard locations of a trailer mounted steady state deflection measurement system's loading wheel and geophones.

 

A different steady state dynamic deflection device uses the electrohydraulic dynamic force generating system. Different models are available with different load magnitudes. One has a peak to peak rated loading from 227 to 1360 kg (500 to 3000 lb); another has a peak to peak rated loading from 454 to 2270 kg (1000 to 5000 lb); a third has a peak to peak rated loading from 454 to 3630 kg (1000 to 8000 lb). Different loading plate sizes are used for various models. Deflections are measured by four velocity transducers. One transducer is located in the center of the loading plate and the other three are attached to a bar, normally at 305 mm (1 ft) intervals. The system can measure the full deflection basin and was widely used in the past due to its reliability. However, the disadvantages include low load levels relative to actual truck loading, a relatively large static preload and susceptibility to measurement errors due to pavement resonance. Standard load wheel/geophone configurations are shown in Figure 2.

Figure 2 - Standard location of an electrohydraulic dynamic force generating system's loading wheel and geophones.

 

The WES 71 kN (16 000 lb) vibrator, developed by the US Army Waterways Experiment Station is contained in an 11 m (36 ft) semi trailer. It uses a 7260 kg (16000 lb) static preload and a dynamic force generator produces a peak to peak loading of around 13 600 kg (30 000 lb). The dynamic load is measured by a set of load cells mounted on the 457 mm (18 in.) loading plate. Deflections are measured by velocity transducers at preselected distances from the loading plate. The entire operation is automated and this device is specifically developed for airfield tests. This is a large unit which is not commercially available at this time. The advantages of this unit include: the large variable loading range of up to 13 600 kg (30 000 lb); variable load frequency ranging from 5 to 100 Hz; and applicability for use on heavy pavements. Disadvantages include: its size; potential susceptibility to measurement errors due to pavement resonance; and its general unavailability attributable to its one of a kind status.

 

Impulse Deflection Equipment
Currently, impulse deflection equipment is the most popular and widely used pavement deflection measurement technology. All impulse type NDT devices produce a transient load to the pavement surface typically lasting 25 to 30 ms. The impulse load is generated by a falling mass from one or more predetermined heights. The resulting load pulse is transmitted to the pavement as a half sine wave. The peak deflections and load magnitude are captured, reported and automatically stored. Testing procedures with impulse load devices are documented in ASTM 4694-96 and ASTM D4695-03. Figure 3 shows a typical time history plot of a falling weight deflectometer load pulse.

Figure 3 - Typical force output from falling weight deflectometers (time from A to B is variable, depending on drop height): A = time at which weights are released; B = time at which weight package makes first contact: C = peak load reached.

 

Impulse load devices can apply loads from 1360 to over 22 700 kg (3000 to over 50000 lb) based on the device used. This equipment has a relatively low preload so its influence on the pavement response is negligible.

Deflections are most commonly measured with velocity transducers (seven or more) which are mounted on a bar and automatically lowered to the pavement surface with the loading plate. One transducer is located in the center of the loading plate and others are located at different distances from the loading plate as shown in Figure 4.

Figure 4 - Typical location of loading plate and deflection sensors for impulse deflection equipment.

 

The Federal Highway Administration's long term pavement performance study specifies deflection sensor spacings at 0, 0.2, 0.3, 0.5, 0.6, 0.9 and 1.5 m (0, 8, 12, 18, 24, 36 and 60 in.) for its testing programs.

The resulting deflections form a "basin" whose depth and shape is used to calculate the material properties of the constitutive pavement layers. These material properties are used to estimate the stress and strain conditions within the pavement structure under the current and expected future traffic conditions. The magnitude of these working stresses and strains are used to estimate the bearing capacity of the pavement and to predict the rate of future deterioration. This information, in turn, is used to assess whether the pavement can meet its expected service life or requires strengthening to meet the anticipated loading conditions.

Typical uses of impulse deflection testing equipment include the following:

• estimation of pavement layer moduli
• overlay design and estimation of remaining life
• load transfer and void determination for concrete pavements
• network level monitoring.

Advantages of impulse deflection testing equipment include high productivity, realistic pavement loading levels, low static preload, rapid data acquisition and the ability to measure and record a deflection basin. However, initial costs for the impulse equipment are higher than their predecessors (static and vibratory devices) and they are more complex in nature.

 

ROLLING WHEEL DEFLECTORS

Background
A rolling wheel deflectometer, as its name suggests, is a device designed to measure and record pavement deflections at highway speeds, thus increasing productivity by an order of magnitude as compared to the falling weight deflectometer. However, the rolling wheel deflectometer is still in the prototype stage and production models are currently not available.

The history of rolling wheel deflectometers dates back to the 1970s. Harr at Purdue University pioneered the measurement method used in the rolling wheel deflectometer devices presented in this tutorial. The technique promised deflection measurements at any speed. A set of four noncontact optical range finders mounted at consistent intervals on a structurally stiff beam form the basic instrument. One optical sensor is mounted near the load tire; the other three are mounted ahead of it and out of the influence of the depression (deflection) basin. These optical sensors operate on the principle of optical triangulation and measure the distance down to the pavement as the rolling wheel deflectometer moves forward. A simple algorithm is used to calculate the pavement deflection as the beam is transported over the pavement.

The deflection measurement procedure first identified by Harr is a two step process. In its simplest form, three equally separated points are measured on the pavement surface ahead of the load wheel. "Measured" in this sense means gaging the distance from a reference datum down to the pavement surface using noncontact optical sensors. When the rolling wheel deflectometer has moved forward, a distance equal to the sensor separation distance, the same three points on the pavement are again measured; but now using the second, third and fourth sensor. In this second step, the load wheel is now located at the rearmost of the three points.

A major problem with the Harr approach is that it required the mounting beam to remain perfectly straight at all times during the measurement process. Therefore, beam bending due to temperature and dynamic loading had to be prevented. Thermal bending could be prevented by controlling the temperature of the beam but the dynamic problem was only solved in the 1990s with a deceptively simple solution: allow the beam to bend, but measure the bending and apply the appropriate corrections. To quantify the beam bending, a laser beam was projected along the long axis of the beam and optical transducers were used to monitor the amount of bending. In 1992, the US government funded the development of a prototype rolling wheel deflectometer using this approach.

 

Prototypes
A prototype version of a rolling wheel deflectometer was developed for testing airfield pavements. This version of a rolling wheel deflectometer continuously measures the maximum pavement deflection near a loaded test wheel while traveling at 6.4 km/h (4 mi/h) along the pavement (although it has been operated at 32.2 km/h [20 mi/h] in recent experiments).

This prototype rolling wheel deflectometer used a series of dual triangulating lasers to determine the maximum deflection. The beam bending is continuously monitored using additional optical alignment sensors mounted on each pavement height sensor. The accuracy (one standard deviation) of the rolling wheel deflectometer pavement height sensor was 20 µm (8 x 10^-3 in).

A parallel rolling wheel deflectometer development effort uses the same "spatially coincident" methodology for measuring pavement deflection. The research and development has been sponsored primarily by the Federal Highway Administration (FHWA) and the Small Business Innovation Research program.

In July 2003, a comprehensive field test was made in College Station, Texas, sponsored by the FHWA and the Texas Department of Transportation. The rolling wheel deflectometer was tested on six pavements representing a range of surface characteristics and deflection levels. A total of 38 individual sections were designated for testing of the rolling wheel deflectometer's effectiveness over a wide range of conditions. Falling weight deflectometer and rolling dynamic deflectometer data were collected over the same sections for comparison to the rolling wheel deflectometer data. In addition, two test sections were fitted with multidepth deflectometers to provide a reference deflection for comparison to the rolling wheel deflectometer. The test results from the rolling wheel deflectometer measurements over six test roads ranging in length from 3.2 to 25.7 km (2 to 16 mi) suggest that the rolling wheel deflectometer technique produced a good repeatability in terms of both deflection magnitude and trends in pavement stiffness. However, the rolling wheel deflectometer experiences a warming up effect prior to stabilization of readings, where the first one or two runs of the repeated measurements show systematically higher deflections than the others.

According to the report prepared for the Texas Department of Transportation and the FHWA, the 16.2 m (53 ft) trailer was custom designed and built specifically for the rolling wheel deflectometer. Its length minimizes pitching of the reference beam, thereby minimizing the laser range needed to accommodate bouncing of the trailer during normal operation. In addition, its natural frequency of 1.45 to 1.8 Hz is low enough that it does not couple with the high frequency vibration of the 7.8 m (25.5 ft) aluminum beam.

 

CALIBRATION OF NONDESTRUCTIVE TESTING EQUIPMENT

Periodic calibration of deflection testing equipment is necessary to obtain accurate pavement deflection data. To address this, the Strategic Highway Research Program and Long Term Pavement Performance program developed a set of standardized falling weight deflectometer calibration procedures and also set up four regional falling weight deflectometer calibration centers. The procedures were subsequently adopted as an American Association of State Highway Transportation Officials provisional standard, "Practice for Calibrating the Load Cell and Deflection Sensors for a Falling-Weight Deflectometer." The calibration centers are located at Austin, Texas; Maplewood, Minnesota; Denver, Colorado; and Harrisburg, Pennsylvania.

 

GROUND PENETRATING RADAR TECHNIQUES

Ground penetrating radar technology is typically used in the following applications:

• pavement layer thickness determination
• subsurface moisture detection
• subsurface void detection
• the detection of concrete pavement deterioration
• discontinuity detection.

Ground penetrating radar transmits short pulses of electromagnetic energy into the pavement from an antenna attached to the survey vehicle. These energy pulses are reflected back to the antenna with an arrival time and amplitude that vary according to the depth and nature of dielectric changes in the underlying material (air/asphalt/base) as shown in Figure 5. The reflected energy captured by the antenna is displayed on an oscilloscope as a radar waveform. A radar waveform consists of a series of reflected energy pulses. The underlying layer properties and thicknesses can be obtained by carefully analyzing the radar waveform.

Figure 5  - Schematic overview of ground penetrating radar techniques.

Figure 6 - Processed ground penetrating radar data.

 

The pavement layer thicknesses and properties can be calculated by measuring the amplitude and arrival times of the waveform peaks corresponding to reflections from the interfaces between the layers. Figure 6 shows the processed pavement layer information from ground penetrating radar testing performed along an asphalt roadway. The peak, A₁ in Figure 5, corresponds to a reflection from the pavement surface. Peaks A₂ and A₃ are reflections from the pavement layer interfaces. Knowing the speed of the wave along with t₁ and t₂, the thickness of the base and subgrades can be calculated. The magnitude of the peaks A₂ and A₃ are proportional to the ratio of the dielectric constants of layers two and three. If the dielectric constants are similar, no peaks will be detected. In most pavements, there is a sufficient difference in dielectric constants of the main pavement layers to accurately measure the various layer thicknesses. However, in some cases, such as a cement stabilized base under a concrete slab, the dielectric constants are not sufficiently different to resolve the thicknesses of the two layers.

Since the radar pulse has its own width, the layers must be thick enough for the reflections from each layer to be clearly resolved. This minimum thickness can be calculated from the radar pulse width (in nanoseconds) and the radar velocity in the medium. For the horn antennas commonly used for this application, this thickness is approximately 63.5 mm (2.5 in.) in asphalt. With ground coupled dipole antennas, such as those commonly used for geotechnical applications, the transmit pulses are two to three times longer (due to ringing) and the thickness resolution is limited to much thicker layers.

To resolve pavement layer properties from radar data, two important assumptions were made: the layers are homogeneous and the layers are nonconductive. These assumptions are not always true. For example, when the layers within the asphalt are not uniform, which may occur due to overlays, intermediate reflections will occur within the asphalt layer. This error can be corrected by recognizing the possibility of layering within the asphalt and by incorporating this layering into the pavement model. The second assumption is generally valid for asphalt but may be less valid for concrete and base material due to higher moisture content. Further details on the pavement layer thickness measurements are presented in a technical paper in this issue.

The benefits of using the ground penetrating radar technique in pavement testing include:

• high productivity - data collection can often be done at highway speeds
• accurate, specific location information for better planning
• rapid data acquisition
• elimination of lane closures, reducing cost and improving worker safety
• collection of hundreds of kilometers of information in a single day
• improvement of falling weight deflectometer determination of elastic modulus and pavement load capacity through providing additional information regarding pavement layer thickness at each falling weight deflectometer drop test location.

 

SPECTRAL ANALYSIS OF SURFACE WAVES

The spectral analysis of surface waves is a relatively new in situ seismic technique for determining pavement layer properties and thicknesses. Testing is performed on the surface, allowing for less costly measurements than with traditional borehole methods. The basic setup of the spectral analysis of surface waves technique is shown in Figure 7.

Figure 7 - Basic setup for spectral analysis of surface waves (GeoVision, 2004).

 

A dynamic source is used to generate surface waves of different wavelengths (or frequencies) that are monitored by two or more receivers at known offsets. Data obtained from forward and reverse configurations are averaged together.

This technique uses dispersion of surface waves to produce a surface wave velocity cross section of the subsurface. As shown in Figure 8, longer wavelengths travel deeper into the subsurface than shorter wavelengths. Thus, different depths of the subsurface can be characterized by generating longer and shorter wavelengths. The velocities of different wavelengths can be determined by calculating the phase difference between two receivers for each wavelength generated.

Figure 8 - Use of different wavelengths to sample different depths (GeoVision, 2004).

 

Surface wave velocities and shear wave velocities are related to each other by Poisson's ratio. Hence, the shear wave velocities can be obtained from surface wave velocities by using inversion methods. Spectral analysis of surface waves can thus be used to map shear wave velocities as a function of depth and chainage, allowing for calculation of the stiffness (maximum shear modulus G_max) of a site.

The operation of the system is simple, being a multichannel seismograph where an impact source like a sledgehammer is used. Tightly spaced shots are generated (typically two times the geophone separation) to cover all the depths of interest. To obtain low frequencies for deeper penetration, a drop weight can be used as an impact source. Data are acquired in a profile mode and shear wave velocity cross sections are produced as a final product.

The dispersion curves can be interpreted using several options which produce different degree of accuracy in the final shear wave velocity profile. A simple empirical analysis can be done to estimate the average shear wave velocity profile. Forward modeling of fundamental mode rayleigh wave dispersion as well as full stress wave propagation can be performed for greater accuracy. A formal inversion scheme may also be used. By incorporating the background information on the site with the above analytical approaches, the resolution of the final profile may be quantified.

The spectral analysis of surface waves offers significant advantages than traditional borehole measurements. The spectral analysis of surface waves is a global measurement where a much larger volume of the subsurface is sampled. The resolution in the near surface (top 7.6 m [25 ft]) is typically greater than with other methods. Significant cost savings can be obtained using spectral analysis of surface waves at sites with favorable surface wave propagation characteristics.

The spectral analysis of surface waves can be used for pavement testing, mapping subsurface stratigraphy, mapping bedrock topography, mapping of subsurface cavities and old mine workings, mapping of low velocity/density zones, mapping of fracture zones in bedrock and calculation of shear wave velocities and site stiffness (G_max).

 

SUMMARY

Nondestructive testing of pavement structures has gained popularity in the recent past. The key information derived from these tests include the following:

• pavement layer thicknesses
• pavement layer structural properties
• void detection under concrete slabs
• load transfer properties across concrete pavement joints.

The measured pavement layer properties are used to determine structural integrity of the pavement structure, assess the remaining structural life of the pavement, determine overlay design, apply load restrictions and so on. It is often preferred to use fast, economical and repeatable pavement test techniques than traditional destructive test techniques. In urban roads, typically there are numerous structures such as gas lines, water mains, cable television and telephone conduits of different diameters at various depths and diameters. The presence of the above makes the use of nondestructive test techniques extremely important.

 

REFERENCES
ASTM International, ASTM D4694-96, Standard Test Method for Deflections with a Falling-weight-type Impulse Load Device, West Conshohocken, Pennsylvania, 1996.

ASTM International, ASTM D4695-03, Standard Guide for General Pavement Deflection Measurements, West Conshohocken, Pennsylvania, 2003.

American Association of State Highway and Transportation Officials, "Benkelman Beam Apparatus," AASHTO T256-77, Washington DC, 1977.

Briggs, R.C., R.F. Johnson, R.N. Stubstad and L. Pierce, "A Comparison of the Rolling Weight Deflectometer with the Falling Weight Deflectometer," Nondestructive Testing of Pavements and Backcalculation of Moduli, Vol. 3, ASTM STP 1375, S.D. Tayabji and E.O. Lukanen, eds., West Conshohocken, Pennsylvania, American Society for Testing and Materials, 1999.

ERES Consultants, "Rolling Wheel Deflectometer (RWD) Demonstration and Comparison to Other Devices in Texas," Champaign, Illinois, ERES Consultants, 2004.

GeoVision, Web site <www.geovision.com>, 2004.

National Research Council, SHRP P-397, Ground Penetrating Radar Surveys to Characterize Pavement Layer Thickness Variations at GPS Sites, Strategic Highway Research Program, Washington DC, National Research Council, 1994.

Shahin, M.Y., Pavement Management for Airport, Roads and Parking Lots, New York, Chapman & Hall, 1994.

Washington State Department of Transportation, WSDOT Pavement Guide, Vol. 2, Olympia, Washington, 1995.

 

* Dynatest Consulting, Inc., Production and Support Center, 13953 US Highway 301 South, Starke, FL 32091; e-mail <rbriggs@dynatest.com>.

 

Copyright © 2004 by the American Society for Nondestructive Testing, Inc. All rights reserved.