Pavement Subgrade, Unbound Materials, and Nondestructive Testing

The integrity of existing highway and airfield pavement is increasingly being evaluated based on the analysis of falling weight deflectometer (FWD) deflection basin data. Measured deflection bowls are matched to deflection bowls computed from multi-layered elastic analysis; the resulting "backcalculated" layer moduli are sometimes blindly used to forge pavement rehabilitation design decisions. However, as has been clearly documented in the literature, the static analysis of dynamic phenomena and the existence or lack of a near surface rigid layer dramatically affects the backcalculated moduli, particularly the subgrade moduli. This paper presents a detailed investigation of soft subgrade moduli via backcalculation from FWD data, triaxial testing, and pressuremeter testing. FWD testing was conducted and data was collected at more than 80 locations along a 1200 m general aviation flexible runway pavement. Deflections were significant, reaching 1.9 mm for an applied stress of 700 kPa. Layer moduli were determined using the widely accepted and Strategic Highway Research Program-recommended MODULUS backcalculation program. For comparison, the subgrade modulus was determined using the pavement pressuremeter and laboratory triaxial testing. Both the pressuremeter and the triaxial test results revealed very soft subgrade soil with moduli ranging from 5 to 47 MPa. Conversely, moduli backcalculated from FWD tests varied considerably depending on the depth to bedrock. Using the shallow depth to bedrock estimated by MODULUS, backcalculated subgrade moduli ranged from 25 to 60 MPa, however, using the actual depth to bedrock measured through site investigation, backcalculated subgrade moduli ranged from 65 to 180 MPa.

Forty laser range finders mounted on a truck are used to assess two transverse profiles. One profile constitutes an unloaded case. It is placed 2.5 m behind the front wheel axle, where the road is little influenced by any wheel load. The other profile, 0.50 m behind the rear axle of the vehicle, constitutes the loaded case. By subtracting the front cross profile from the corresponding rear one, the "transverse difference profile" is assessed. The difference is assumed to vary with the stiffness of the road. A high sampling rate is needed for filtering the macro texture of the pavement. Averaging profiles over some section length should also mitigate effects of vehicle and instrument vibration. It is assumed that the slope of the deflection basin near the load is influenced by near surface elastic properties. The total area and the slope of the deflection profile far from the load can be attributed to layers deep in the structure. Thus, using simple area parameters the influence by the subgrade stiffness on the deflection basin caused by the load can be estimated. During the spring and summer of 1998 the first major test program was carried out with the new rolling deflection tester. The present paper compares deflection basins assessed with those calculated with a finite element method. Many different subgrade conditions were tested and the results from these tests are presented. Further, the results are compared to processed data achieved from falling weight deflectometer testing.

The SASW test is a seismic technique for in situ evaluation of elastic moduli and layer thicknesses for layered systems, such as pavements. The objective of the SASW test is to obtain the experimental dispersion curve, and then through an inversion or backcalculation procedure to obtain the elastic modulus profile. The inversion process is a complex process that requires a significant computational effort and often requires operator's intervention. An artificial neural network was developed, that backcalculates accurately in real time pavement profiles from the SASW test collected data. The network is developed using synthetic dispersion curves from a numerical simulation of the SASW test. The most important feature of the developed network is that it consists of several neural network models used in evaluation of the thickness and elastic modulus of each individual layer. This results in a significant improvement in accuracy of evaluation of pavement properties. In comparison to the previous ANN models, accuracy can be attributed to the individual layer approach and to the use of a data transformation algorithm. The stiffness matrix approach used in generation of synthetic dispersion curves is described.

The recent adoption and use of mechanistic concepts in pavement design made it possible to more realistically characterize the unbound aggregates component of the pavement by incorporating the new advances in characterization, i.e., anisotropic, stress path dependent behavior, and accounting for the most damaging field conditions. This paper mainly focuses on the use of a unique triaxial testing machine, referred to as University of Illinois FastCell (UI-FastCell), for determining in the laboratory the stress-induced anisotropic resilient properties of twelve aggregates with varying material types and properties. In the selection process, consideration was given to both "good" and "poor" performing granular base/subbase materials obtained from seven different states. With the applied AASHTO T294-94 stress states conveniently pulsed either in the vertical or horizontal direction on the same specimen, the anisotropy (directional dependency) of resilient moduli are adequately determined in the laboratory for the compression and extension type dynamic loadings. Nonlinear stress dependent models are developed to characterize the resilient moduli under these extreme loadings. The moduli obtained under extension stress states are in general lower than the ones computed from compression tests. Such lower moduli would result in higher critical pavement responses as obtained from the mechanistic analysis of a conventional flexible pavement and be associated with the reduced pavement life eventually. The most significant stress history effect is observed for the materials having moderate amount of fines. Finally, anisotropic (horizontal to vertical) modular ratios corresponding to horizontal (extension) and vertical pulsing (compression) conditions are shown to either increase or decrease with increasing applied dynamic stresses depending on the material properties, i.e., quality of an aggregate.

The results from static load-based backcalculation programs indicate sensitivity to the assumed depth to bedrock. In this study, a mechanistic method has been developed for estimating the apparent depth to the stiff layer using the Falling Weight Deflectometer (FWD) data. This predicted depth can be used in theoretical pavement models to assure that the fixed model boundary has an insignificant effect on the estimated surface deflection. The approach is based on utilizing Three-Dimensional Finite Element (3D FE) to simulate pavement response to the FWD impact load. This approach accounts for the dynamic nature of the FWD load, friction at the layers' interfaces, and the 3D geometrical characteristics of pavement structures. After adjusting the model parameters so that the 3D FE-generated deflection basin matches the experimentally measured one, the decay of displacements and stresses in the subgrade layer thickness is investigated along the axis of the FWD load application. The apparent depth to bedrock is defined as the depth at which the subgrade stress decays to approximately 10 percent of its value measured on the subgrade surface (Analogous with the settlement calculation of shallow foundation). Investigation of the effect of an assumed subgrade layer thickness on the 3D FE pavement model response to the short duration FWD-impact load, reveals that as the subgrade layer thickness is increased from 1.0 m (3.3 ft) to 3.66 m (12.0 ft), the change in surface deflections is less than 3 percent. Thus, for the objective of minimizing computation time and memory requirements of 3D FE pavement models, a subgrade layer thickness of around 1.83m (6 ft) will be sufficient to produce satisfactory results.

The resilient modulus is a basic soil property governed by the elastic theory, and is a measure of the recoverable strain for a small, repetitive load. It is a required input in the AASHTO method of flexible pavement design. The resilient modulus of subgrade soils is a required input parameter for the AASHTO method of flexible pavement design and is also used to compute the modulus of subgrade reaction for the AASHTO rigid pavement design equations. This study analyzed subgrade resilient moduli, backcalculated from the Falling Weight Deflectometer deflection test results, on nine pavement subgrade projects of the Kansas Department of Transportation (KDOT). Laboratory resilient modulus test results were also reviewed for the fine-grained soils on these projects. The results show that the average "breakpoint" deviator stress for the fine-grained soils that displayed typical bi-linear behavior was 46 kPa. The design resilient modulus corresponding to this deviator stress value could be used in the instances when the "breakpoint" deviator stress is not distinct graphically. The average ratio of the laboratory-determined to the backcalculated resilient moduli was found to be 0.33. Backcalculated resilient moduli using the AASHTO algorithm and a nonlinear elastic analysis program were nearly identical for the soils that had slightly nonlinear characteristics. Results from either approaches can be used in design.

The heavily instrumented Virginia Smart Road project provided the opportunity to test various hypotheses on pavement nondestructive testing using falling weight deflectometer (FWD). This was achieved by analyzing the deflections obtained on top of each layer as the road was being constructed. Two loading plate sizes were used at the subgrade and aggregate subbase surfaces. Five different loadings were used in each testing. This allows an accurate backcalculation of insitu resilient modulus of each layer. Such results may be used to calibrate future FWD measurements. The study found that using the surface modulus to characterize the subgrade might sometimes be misleading. In addition, characterization of the deflection basin using deflections alone may not provide accurate results. Using the ratio between the deflection basin area centroid coordinates may provide a reasonable alternative. The distribution of pressure resulted from FWD loading was successfully measured by embedded pressure cells in the HMA base layer.

Experimental investigations of nonlinear shear viscosity of subgrade soils under periodic loading have been conducted. Unsaturated subgrade soil samples were prepared to conduct triaxial shear tests under repeated loading that was used to simulate the cyclic stress induced by vehicles. Investigations included three aspects. First, a new nonlinear viscous model has been introduced to describe the relation of shear stress and shear strain rate. This relation has been reduced from the constitutive law of nonlinear poroviscosity suggested by Li (1994) as well as Li and Helm (1995 and 1997). Second, related constitutive parameters have been investigated and calibrated with experimental results from triaxial shear tests. In the present paper, nonlinear viscous parameters are assumed to be a function of deviatoric strain and loading repetitions. Third, both viscous and elastic behavior of soils associated with the deviatoric strain and the cyclic loading repetition has been studied. The correlation between shear viscosity and resilient modulus has been discussed as well. This is the first attempt to find any correlation between soil viscosity and elasticity through laboratory investigations. In contrast, in traditional numerical computations (e.g., the finite element method), the extra viscous term added to the governing equation is assumed to be related to system mass and rigidity.

A number of dynamic rigid pavement models rely on a damped Winkler-type foundation for characterizing the soil support. While the static subgrade modulus can be determined from existing backcalculation procedures, researchers have typically picked an arbitrary value for the damping ratio. This approach cannot be used for a realistic characterization of a specific subgrade soil condition in a particular site. In fact, for certain site conditions, it may lead to a completely erroneous characterization of foundation support. This paper presents a simple and rational method for backcalculating the dynamic subgrade stiffness and damping coefficients from Falling Weight Deflectometer (FWD) deflection basins. The method can also be used to detect a stiff layer underneath the pavement system. The method consists in first decomposing the transient deflection signal of each sensor into a series of harmonic motions using Fast Fourier Transform (FFT) algorithm. Then, for each frequency of interest, the real and imaginary components of the displaced volume underneath the slab are calculated from the complex deflection basin. The dynamic force- displacement relationship is decomposed into real and imaginary parts, leading to a simple system of equations that can be easily solved for k and c. The method was verified for different soil profiles using both analytical and experimental results. The analytical verification used results from 3-D finite element analysis. The experimental verification used results from deflection measurements on airfield and highway pavements obtained with the WES vibrator and MDOT Falling Weight Deflectometer, respectively. The results showed very good agreement.

Temperature, subgrade moisture content, and Falling Weight Deflectometer (FWD)-deflection data were collected monthly on four asphalt pavement test sections in Kansas for a year. The subgrade moduli were backcalculated using the elastic layer theory. It was found that for almost all sites, the monthly variation in subgrade moisture content was not very significant over the seasons. The patterns of subgrade response, in terms of subgrade moduli versus time, simulated sine-shaped forms signifying a possible temperature effect. Higher variabilities across the site were associated with the extreme temperature conditions, usually very low or high average pavement temperatures. In all cases, the measured precipitation was nominal thereby excluding this climatic variable as a major factor. Extreme test temperatures, both high and low, result in higher variation of measured deflections and subsequently, backcalculated subgrade moduli across a site. Thus, some variabilities in backcalculated subgrade moduli can be minimized by conducting FWD tests in a moderate temperature regime. The analysis of variance (ANOVA) indicated that both seasonal and site variabilities can be significant. After correction for temperature, variations in deflections and moduli become approximately equal to the site variabilities which was also confirmed by ANOVA.