Stefan A Romanoschi

Characterization of Asphalt Concrete Layer Interfaces

A new constitutive model for the asphalt concrete layer interface is proposed. Direct shear tests at four levels of normal load and three temperatures were performed on two types of asphalt concrete layer interface: with and without a tack coat. The shear stress-displacement curves determined in these tests were used to derive the constitutive model, as the tangential and normal stresses at the interface are decoupled. In the proposed model, the shear stress and displacement are proportional until the shear stress equals the shear strength and the interface fails. After failure, a friction model may be used to represent the interface condition. Three parameters were considered to completely describe the interface behavior: the interface reaction modulus K, which is the slope of the shear stress-displacement curve; the shear strength Smax; and the friction coefficient after failure μ. For the interface with a tack coat, K and Smax are not affected by the normal stress level, but they are affected for the interface without a tack coat. All three parameters of the constitutive model are temperature dependent. A testing configuration for determining the shear fatigue behavior of the interface is also described. The fatigue tests indicated a linear increase of the permanent shear displacement with the number of load repetitions, the rate of increase being higher for higher stresses. The fatigue test can be used for a comparative evaluation of the durability of different types of interfaces.

Accelerated pavement testing evaluation of the structural contribution of full-depth reclamation material stabilized with foamed asphalt

Research was conducted to determine the effectiveness of the use of foamed asphalt-stabilized reclaimed asphalt pavement from full-depth reclamation (FAS-FDR) as base material for flexible pavements. The experiment, conducted at the Civil Engineering Infrastructure Systems Laboratory of Kansas State University, consisted of constructing four pavements—one with a 9-in. conventional Kansas AB-3 granular base and one each with 6, 9, and 12 in. of FAS-FDR—and subjecting them to a full-scale accelerated pavement test. All four pavement sections were loaded with 500,000 axle load repetitions, at room temperature and under moderate moisture levels in the subgrade soil. The measured stresses and strains as well as the permanent deformation (rutting) observed on the pavement sections indicated that FAS-FDR can be used successfully as a base material. The measured rut depths and compres-sive vertical stresses at the top of the subgrade suggest that a 1-in. FAS-FDR base shows performance equivalent to that of a 1-in. conventional Kansas AB-3 granular base. The effective structural number computed from the falling weight deflectometer tests on the as-constructed pavements showed that average structural layer coefficient for the FAS-FDR base material was 0.18.

Effects of interface condition and horizontal wheel loads on the life of flexible pavement structures

The effects of interface condition and horizontal wheel loads on the life of flexible and semirigid pavements were determined. The methodology consisted of implementing a previously derived interface constitutive model into the ABAQUS finite element program to compute the stresses and strains in typical flexible and semirigid road structures. The Shell transfer functions for fatigue cracking and terminal serviceability were used to estimate the life of the two pavements. The study revealed that the horizontal loads acting at the pavement surface lead to dramatically increased tensile strains at the top and bottom of the wearing course and at the top of the binder course. This may justify the initiation of cracking at the surface of the pavement and not at the bottom of the asphalt layer, as generally assumed. For semirigid pavements, the condition of the wearing-binder course interface affects the strains in the wearing course, whereas the condition of the binder-base interface affects the horizontal strain field in the binder layer more as well as the vertical strains at the top of the subgrade. For flexible pavements, the condition of the interface between the wearing and binder courses dramatically changes the strain field in the wearing and binder layers and increases the vertical strains at the top of the granular base and subgrade layers. The cumulative effect of the interface condition and horizontal forces acting at pavement surface is expressed by a dramatic reduction in pavement life, especially for the semirigid pavement.

Soil Stiffness Evaluation for Compaction Control of Cohesionless Embankments

Mechanistic pavement design procedures based on elastic layer theory require characterization of pavement layer materials including subgrade soil. This paper discusses the subgrade stiffness measurements obtained from a new compaction roller for compaction control on highway embankment projects in Kansas. Three test sections were compacted using a single, smooth steel drum intelligent compaction (IC) roller that compacts and simultaneously, measures stiffness values of the compacted soil. Traditional compaction control measurements such as, density, in-situ moisture content, stiffness measurements using a soil stiffness gage, surface deflection tests using the light falling weight deflectometer (LFWD) and falling weight deflectometer (FWD), and penetration tests using a dynamic cone penetrometer (DCP), were also done. The results show that the IC roller was able to identify the locations of lower soil stiffness in the spatial direction. Thus, an IC roller can be used in proof rolling. IC roller stiffness showed sensitivity to the field moisture content indicating that moisture control during compaction is critical. No universal correlation was observed among the IC roller stiffness, soil gage stiffness, back-calculated subgrade moduli from the LFWD and FWD deflection data, and the California bearing ratio obtained from DCP tests. The discrepancy seems to arise from the fact that different pieces of equipment were capturing response from different volumes of soil on the same test section.

Intelligent Compaction Control

The Intelligent Compaction Control (ICC) consists of continuous compaction control/monitoring compaction using rollers with adjustable compaction energy ( amplitude, frequency, and roller speed ). In ICC, a number of parameters are measured: displacements/amplitude of the roller (up and down) using the drum mounted accelerometer, frequency, roller speed, and various relative bearing capacity or equivalent stiffness/density values. This paper gives an introduction to ICC and application of ICC for highway embankment projects in Kansas. Three test sections on two routes were compacted using a Bomag Variocontrol (BVC) intelligent roller that produces real time stiffness values of compacted soil. Traditional compaction control measurements included density testing using a nuclear gage, moisture measurements using a speedy moisture tester, and soil bearing capacity measurements using a Dynamic Cone Penetrometer (DCP). The results showed that the intelligent compaction (IC) roller continuously measured the stiffness of soil under compaction and thus, was able to identify locations with lower stiffness in the spatial direction. In general, density increased with multiple passes of the IC roller. The IC roller stiffness was fairly sensitive to the moisture content and the percent compaction obtained in the field. Poor correlation was observed between the BVC stiffness and the CBR values calculated from the DCP results.

First Findings from the Kansas Perpetual Pavements Experiment

To investigate the suitability of the perpetual pavements concept for Kansas highway pavements, the Kansas Department of Transportation (KDOT) constructed four thick, flexible pavement structures on a new alignment on US-75 near Sabetha, Kansas. They were designed to have a perpetual life and have layer thicknesses close to those recommended by KDOTs structural design method for flexible pavements, which is based on the 1993 AASHTO Design Guide. To verify the approach of designing perpetual pavements on the basis of an endurance strain limit, the four pavements were instrumented with gauges for measuring the strains at the bottom of the asphalt base layers. Seven sessions of pavement response measurements under known vehicle load were performed between July 2005 and October 2007, before and after the pavement sections were opened to traffic. The analysis of the strain data indicated that, even during hot summer days, the strains of all four test sections were smaller than the endurance limit of asphalt–concrete. As expected, the strains were affected by the temperature in the asphalt layers and the speed of the loading vehicle. The analysis of the strain signals revealed that the transverse strain under the front axle did not recover completely before the arrival of the rear axles, a situation causing the accumulation of dynamic transverse strain to values higher than those of the corresponding longitudinal strains. A comparison between the measured response and that predicted by a linear-elastic model indicated that the predicted transverse strains were close to half the corresponding measured dynamic transverse strains, while the predicted longitudinal strains were close to twice the measured dynamic longitudinal strains. Furthermore, the predicted vertical stresses at the top of the subgrade layer were close to five times the measured stresses.

Development of Traffic Inputs for New "Mechanistic–Empirical Pavement Design Guide": Case Study

Vehicle classification and axle load data are required for the structural design of new and rehabilitated flexible and rigid pavements with the new Mechanistic–Empirical Pavement Design Guide (MEPDG) developed under NCHRP Project 1–37A. The axle load spectra are determined from traffic data collected at weigh-in-motion (WIM) stations, and vehicle count and class data are recorded by vehicle classification stations. Some preliminary results are presented for an extensive traffic data-processing effort conducted to develop traffic inputs required by the MEPDG to design pavements in New York State. The data collected by classification and WIM sites from 2004 to 2009 were processed with the TrafLoad software developed in NCHRP Project 1–39. The discussion focuses on the variability of the major traffic input variables required by the MEPDG, as obtained from data collected in New York State, and on the differences between the data obtained from individual stations, state average values, and the default values recommended by the MEPDG, where applicable. The effect of variability of the major traffic input variables on the performance predicted by the MEPDG for a typical flexible pavement structure is also discussed.

EVALUATION OF PROBABILITY DISTRIBUTION FUNCTION FOR THE LIFE OF PAVEMENT STRUCTURES

Determination of the probability distribution function for the time to failure is essential for the development of pavement life models, because the probability distribution function reflects the variability in pavement degradation. The pavement life and failure time are associated with the number of equivalent standard axle load applications for which the degradations reach a critical level. When the critical degradation level is reached, maintenance and rehabilitation work needs to be done to improve pavement condition. Research was undertaken to identify the appropriate statistical models for determination of the probability distribution function for the time to failure of pavement structures. The study used the rutting data collected on a test lane at the first full-scale accelerated pavement test in Louisiana. The research indicated that closed-form solutions or Monte Carlo algorithms can be used when the degradation models have a known form. The bootstrap algorithm can be used to determine the confidence intervals for probability of failure at a given time. If the form of the degradation model is not known, the survival analysis method based on censored observations must be used. The methods can be used not only for rutting life models but also for other pavement life models: cracking initiation time, cracking life, roughness, and serviceability lives.

Performance and Failure Modes of Louisiana Asphalt Pavements with Soil-Cement Bases Under Full-Scale Accelerated Loading

The performance and failure modes of conventional soil-cement base asphalt pavements and alternative pavements have been investigated under accelerated loading. Surface cracking was evaluated in terms both of the crack rate (in meters per square meter) and the AASHTO Class 1, 2, and 3 classifications. Pavement structural capacity was evaluated in terms of falling weight deflectometer (FWD) deflection-based back-calculated moduli of asphalt layer and soil-cement bases. Analysis showed that there was no significant difference in pavement performance among the soil-cement base pavements, although the cement content, strength, and construction procedures for the soil-cement bases were different. A significant improvement in pavement fatigue life was found for the “inverted” pavement structure, in which a stone crack relief layer was placed between the soil-cement base and the asphalt surfacing. The fatigue life of the inverted pavement was five to six times longer than that of pavements without the stone layer. Different failure modes were found between the inverted pavement and the others, and the corresponding failure mechanisms were analyzed on the basis of the observations from the pavement postmortem, when loading was terminated. Analysis of the FWD deflection data indicated that under accelerated loading there was a significant decrease in asphalt and soil-cement moduli for the inverted pavement, although the same trend was not observed on the other soil-cement bases. The crack data were used to evaluate two pavement fatigue life prediction models. It was found that the current models are not applicable to the soil-cement base pavement with a stone crack relief layer.

Perpetual pavement temperature prediction model

Structural capacities of flexible pavements are determined from surface deflection measurements. These deflections must be corrected to a standard load and/or a reference pavement temperature. A number of models are available to predict pavement temperature, but they may not be applicable to perpetual (thicker) asphalt pavements. Mid-depth pavement temperature was measured in six sessions on four perpetual pavement sections in Kansas. Data from five sessions were used to develop the prediction model based on four independent variables. Data from the last session were used to validate it. Predicted mid-depth pavement temperatures from the new model and three other models were compared with the measured mid-depth pavement temperature. Sensitivity of the model to changes in all independent variables was also investigated. The effect of mid-depth pavement temperature on the centre deflection of the falling weight deflectometer was also studied. The prediction model developed in this study yields mid-depth pavement temperature that is closest to the measured mid-depth temperature. It also results in lowest bias in terms of centre deflection. Predicted mid-depth pavement temperature is most sensitive to the time of day when measurements are made and least sensitive to the layer mid-depth thickness.