Bottom-up Cracking Research Articles

Analysis of load-induced top-down cracking initiation in asphalt pavements using a two-dimensional microstructure-based multiscale finite element method

Load-induced top-down cracking is one of the major types of asphalt pavement deterioration; however, its initiation mechanisms have not been fully understood so far, which makes it very difficult to effectively consider this failure pattern in the pavement design procedures. To address this issue, the present study developed a two-dimensional microstructure-based multiscale finite element model, in which material properties on two physical length scales, i.e., the local (mixture level) and global (pavement level) scales, were incorporated in the computation and linked through a homogenization process. A digital image processing (DIP) technology was employed to develop the two-dimensional local-scale representative volume element (RVE) model that considered the realistic heterogeneous microstructure of asphalt concrete (AC), and a bilinear cohesive zone model was applied to simulating the local-scale damage initiation and evolution in the RVEs. Two typical pavement structures, with cement-treated base (CTB) and granular base (GB) respectively, were taken into account to interpret the influence of the global-scale pavement configurations on top-down cracking performance. The results showed that the significant near-surface transverse tensile stress just outside the tire edge could be the primary cause of the top-down cracking. For the pavement with CTB, the top-down cracking was the predominant type of fatigue failure, whereas for the pavement with GB, the bottom-up cracking was the main pattern of fatigue failure. Besides, as the temperature increased, more damage was induced under the same traffic loading due to the reduced tensile strength of AC. It was also found on the local scale that the significant tensile stress within the mortar matrix phase probably acted as the driving force of the microcrack initiation and propagation and the effects of the shear traction on the damage evolution in the AC layer increased with the temperature.

Evaluating Dynamic Modulus for Indiana Mechanistic-Empirical Pavement Design Guide Practice

After the implementation of the Mechanistic-Empirical Pavement Design Guide (MEPDG) in Indiana, an overall evaluation of the stiffness characteristics of local AC mixtures and the ability of level III MEPDG predictive equations to estimate dynamic modulus (E*) with local mixtures was required. Therefore, the primary objectives of this study were to identify significant differences among Indiana asphalt mixtures, to evaluate the performance of commonly used E* predictive models, and to assess the influence of level III E* input on the pavement design life of typical pavement structures. It was found that Indiana mixtures do not show extensive variability among mixtures having the same nominal maximum aggregate size. When conducting a statistical analysis to group asphalt mixtures having similar characteristics, few mixtures were left out of the groups. In general, it was observed that mixtures having Ndes equal to 75, showed the lowest E* values along the entire frequency range. The Witczak 1-37A showed the most accurate and less biased E* predictions for Indiana mixtures. It showed the highest R2, and the least deviation from the measured E* values. However, predicted E* input values produced higher levels of pavement distress compared with measured E* values, indicating general overprediction. Besides, using level III (predictive) rather than level I (measured) E* input values can influence the pavement thickness design due to the functional performance (i.e., the International Roughness Index (IRI)). When a structural performance (i.e., bottom-up cracking) was taken into consideration, no influence of the E* input type on the design AC layer thickness was observed.

Mechanistic-Empirical design of perpetual pavement

Perpetual pavements use multiple, durable asphalt layers to produce a safe, smooth, long-lasting road. The pavement design begins with a strong, yet flexible bottom layer that resists tensile strain caused by traffic, and thus stops cracks from forming at the bottom of the pavement. Four perpetual pavement sections were designed and constructed in 2005 and currently are scheduled for rehabilitation for top-down surface cracking. These cracks have been confirmed by full-depth cores. This study revisited the design of these perpetual pavement sections using now available PerRoad and AASHTOWare Pavement ME Design software to investigate whether the design assumptions were correct. Fatigue lives of these sections were also evaluated based on bottom-up cracking. Layer moduli were backcalculated from the falling weight deflectometer (FWD) deflection data using the EVERCALC software. Tensile strains at the bottom of the asphalt layer were then computed. Results showed that predicted strain values derived from the mechanistic response models were lower than the tensile strains computed using moduli from the FWD deflection data. The required asphalt thickness from PerRoad was relatively higher than the asphalt thickness obtained from the AASHTOWare Pavement ME Design software over a 50-year design period. The study identified the most successful perpetual pavement design as the one with a rich bituminous mix in the base layer.

Cold Central Plant Recycled Asphalt Pavements in High Traffic Applications

Cold central plant recycling (CCPR) is gaining wider use in the U.S. for rehabilitating existing asphalt pavements or for new construction. Although it is used widely in lower traffic volume situations, CCPR use in high volume pavements remains an open question when considering its structural capacity and expected performance. A project completed in 2011 on I-81 in Virginia indicated CCPR may be suitable for high-volume traffic applications and was further evaluated with the construction of three CCPR test sections at the National Center for Asphalt Technology Test Track in 2012. These sections are now approaching 20 million equivalent single axle load applications and this paper documents their structural and surface performance thus far. The structural characterization indicates healthy pavements with no significant increases in measured pavement response or decreases in backcalculated moduli over time. Performance has been excellent with no cracking observed on any section, rut depths less than 0.3 inches and ride quality that has remained almost unchanged. Perpetual pavement analyses were also conducted and found that the section with a cement-stabilized base layer supporting the CCPR layer met the criteria and is likely perpetual. The other two sections, without the cement-stabilized base, did not meet the criteria and may develop bottom-up cracking. Data from the I-81 and Test Track sections enabled the Virginia Department of Transport (VDOT) to proceed with a design-build project on I-64 that will feature CCPR with a cement-stabilized base and full-depth reclamation (FDR). It is estimated that nearly 170,000 tons of reclaimed asphalt pavement will be used with over $10 million in savings.

Effects of layer interface conditions on top-down fatigue cracking of asphalt pavements

Different types of responses have been considered in the past as the driving forces of the top-down cracking (TDC). The horizontal tensile strain at the surface is the most appropriate response of pavement structures for analysing TDC. In computing the horizontal strains, most of the mechanistic design methods of asphalt pavement consider that pavement layers are fully bonded and bottom-up cracking is the critical type of fatigue distress. The bonding condition between pavement layers is one of the significant factors influencing pavement performance. In this research, the effects of interface bonding conditions on top-down fatigue cracking are investigated based on the horizontal tensile strain at the surface. Analysis procedures are conducted using of Lucas algorithm and integration, summation and extrapolation (ISE) methods to determine the horizontal strain responses of asphalt concrete (AC) layer at the surface accurately. The results reveal that the AC layer thickness, type of base layer material and temperature affect the horizontal strains at the surface and debonded locations significantly. When the layers of thick pavement with cement-treated base (CTB) are fully bonded, the TDC is a dominant type of distress at different temperatures analysed, while for the thick pavement with granular base (GB) the surface-initiated cracking occurs only at high temperature. Due to the interface debonding in the thick pavement, the maximum horizontal strains vary at different locations and failure may appear simultaneously at the top and bottom of AC layers or only at one of those regions. For the thin pavement structure, the layer interface bonding conditions hardly influence the mechanistic response at the surface and the pavement with GB is more vulnerable to bottom-up cracking, whereas the likelihood of TDC increases at high temperature for the thin pavement with CTB.

Sensitivity quantification of airport concrete pavement stress responses associated with top-down and bottom-up cracking

The Federal Aviation Administration’s (FAA’s) rigid pavement design standard employs the NIKE3D-FAA software to compute critical pavement responses of concrete airport pavement structures. NIKE3D-FAA is a modification of the original NIKE3D three-dimensional finite element analysis program developed by the Lawrence Livermore National Laboratory (LLNL) of the U.S. Department of Energy, and is currently used in the FAA’s FAARFIELD program. This study evaluated the sensitivity of NIKE3D-FAA rigid pavement responses with respect to top-down and bottom-up cracking. The analysis was conducted by positioning a Boeing 777-300ER (B777-300ER)aircraft at different locations (interior, corner, and edge of slab) as baseline while varying other NIKE3D-FAA inputs, including rigid pavement geometric features, mechanical properties of paving and foundation materials, equivalent temperature gradient and thermal coefficient of Portland Cement Concrete (PCC) layers. Several sensitivity charts were developed by examining the sensitivity of critical pavement responses to each input variation. Sensitivity evaluations were performed using a normalized sensitivity index (NSI) as the quantitative metric. Using such sensitivity evaluation, the most significant NIKE3D-FAA input parameters for generating an effective synthetic database that will lower computational cost for future modeling developments were identified.

A new approach for damage detection in asphalt concrete pavements using battery-free wireless sensors with non-constant injection rates

This study presents a novel approach for detection of bottom-up cracking in asphalt concrete (AC) pavements using self-powered wireless sensor (SWS) with non-constant injection rates. The performance of the sensors was evaluated through numerical and experimental studies on an asphalt concrete specimen under three-point bending configuration. Damage was introduced by making notches with different sizes at the bottom of the specimen. Different 3D finite element (FE) models were developed using ABAQUS to generate the sensor output data for different damage states. Thereafter, the laboratory tests were carried on the same specimen to validate the performance of the proposed damage detection approach. Polyvinylidene fluoride (PVDF) piezoelectric film was used to harvest the strain energy from the host structure and empower the sensor. In order to protect the embedded sensor, an H-shape packaging system was designed and tested. The results indicate that the progression of bottom-up cracks can be accurately detected using the proposed self-powered sensing system.

A self-powered surface sensing approach for detection of bottom-up cracking in asphalt concrete pavements: Theoretical/numerical modeling

This study presents a surface sensing approach for detection of bottom-up cracking in asphalt concrete (AC) pavements. The proposed method was based on the interpretation of compressed data stored in memory cells of a self-powered wireless sensor (SWS) with non-constant injection rates. Different 3D finite element (FE) models of an AC pavement were developed using ABAQUS to generate the sensor output data. A realistic dynamic moving load was applied to the surface of the pavement via DLOAD subroutines developed by FORTRAN. A network of sensing nodes was placed at the top of the AC layer to assess their sensitivity to the progression of bottom-up cracks. Several damage states were defined using Element Weakening Method (EWM). A linear-viscoelastic behavior was considered for the AC layer. In order to detect the damage progression, several damage indicator features were extracted from the data acquisition nodes. The damage detection accuracy was improved through a data fusion model that included the effect of group of sensors. The proposed fusion model was based on the integration of a Gaussian mixture model (GMM) for defining descriptive features, different feature selection algorithms, and a robust computational intelligence approach for multi-class damage classification. Furthermore, an uncertainty analysis was carried out to verify the reliability of the proposed damage detection approach. The results indicate that the progression of the bottom-up cracks can be accurately detected using the developed intelligent self-powered surface sensing system.

Investigation of mechanisms of top-down fatigue cracking of asphalt pavement

Various types of responses have been selected in the past as the driving forces of the top-down cracking. However, the most appropriate candidate for top-down cracking is the horizontal tensile strain at the pavement surface, considering the horizontal tensile strain at the bottom of asphalt layer has been historically used as the critical response for bottom-up cracking in the ME design. In this study, the mechanisms of top-down cracking are investigated based on the horizontal tensile strain at the surface. A multi-layered elastic computer code, APRA, was utilized to compute the horizontal strains at the pavement surface. APRA uses the Lucas algorithm and integration, summation and extrapolation methods to accurately determine the surface responses. The results show that traffic loading may induce significant horizontal tensile strain in the transverse direction at the surface, which is a major cause of the top-down cracking. The critical location of the tensile strain is just outside the tyre edges. It is pointed out that although the newly developed Mechanistic-Empirical Pavement Design Guide (MEPDG) uses the horizontal strain at the surface for the prediction of top-down cracking, the analysis locations adopted in the MEPDG may not be able to identify the maximum tensile strain. It is observed in this study that the maximum tensile strain at the surface increases rapidly with temperature. For the pavement with cement-treated base, the top-down cracking is the dominant type of fatigue cracking, while the bottom-up cracking is unlikely due to the strong support from the base. For the pavement with granular base, the bottom-up cracking is more likely to occur at medium or low temperatures, while the cracking is more likely to initiate from the top at high temperatures. The likelihood of top-down cracking occurrence increases with increasing temperature and asphalt layer thickness.

Fatigue Life Prediction of High Modulus Asphalt Concrete Based on the Local Stress-Strain Method

Previously published studies have proposed fatigue life prediction models for dense graded asphalt pavement based on flexural fatigue test. This study focused on the fatigue life prediction of High Modulus Asphalt Concrete (HMAC) pavement using the local strain-stress method and direct tension fatigue test. First, the direct tension fatigue test at various strain levels was conducted on HMAC prism samples cut from plate specimens. Afterwards, their true stress-strain loop curves were obtained and modified to develop the strain-fatigue life equation. Then the nominal strain of HMAC course determined using finite element method was converted into local strain using the Neuber method. Finally, based on the established fatigue equation and converted local strain, a method to predict the pavement fatigue crack initiation life was proposed and the fatigue life of a typical HMAC overlay pavement which runs a risk of bottom-up cracking was predicted and validated. Results show that the proposed method was able to produce satisfactory crack initiation life.

Estimating fatigue endurance limits of flexible airfield pavements

In perpetual pavements, damage from bottom-up cracking can be limited to the top surface lift through using a very thick surface layer or a binder-rich intermediate layer. This can be attained by maintaining tensile strains at the bottom of the hot-mix asphalt (HMA) layer below a certain value known as the fatigue endurance limit (FEL). This paper presents a method for estimating a strain-based FEL for flexible airfield pavements. The proposed method is based on the concept that a 50% reduction in HMA layer modulus would indicate initiation of fatigue cracking. Falling weight deflectometer (FWD) testing results, collected from National Airport Pavement Test Facility (NAPTF) flexible pavement sections, were analyzed to determine at which loading pass each section had a 50% reduction in HMA layer modulus (Nf50). NAPTF tensile strain data were also used to determine the tensile strain at Nf50 for each pavement section by averaging the peak tensile strains. The proposed approach was validated by comparing its results to those obtained using a common FEL estimation model known as the rate of dissipated energy change (RDEC) model. To further verify the results of the proposed approach, peak tensile strain was plotted vs. number of loading cycles for all sensors. Using these plots, the peak tensile strain at which the variability in the strain gauge data increased was used as an estimate of a possible FEL. The Nf50 tensile strains estimated using the proposed method were comparable to the values determined from RDEC and variability approaches.

Advancing the design and construction quality control of flexible pavements

The paper presents several advanced features that can be used in the process of pavement design and construction quality control. The models for predicting the pavement performance in the latest version of mechanistic empirical pavement design guide form the basis of the proposed system. In order to add the new features, it was necessary to make simplifications concerning the moisture distribution in the unbound materials and the ageing/oxidation of the upper asphalt concrete (AC) layers. The add-ins to the program include: (1) An analytical reliability analysis where the variability of the performance prediction components is computed from the variability of the pavement parameters; (2) An improved simulation of the temperature distribution in the AC layers during the design period; (3) An option for adding an overlay during the design period, to simulate maintenance or stage construction and (4) The possibility to use a 2D finite element analysis with nonlinear and cross-anisotropic materials, in addition to the well-established linear elastic analysis. The proposed system runs every hour of the design period and computes the fatigue bottom-up cracking, rutting in all pavement layers and international roughness index. It runs very fast, less than one minute in the linear analysis and less than 30 min in the nonlinear case. Few cases are presented to illustrate the benefits of the above add-ins.

3D modelling of tyre-pavement contact pressure

Pavement design assumes uniform contact loading because of bottom-up traditional cracking (far from loading), but nowadays more and more surface damage (top-down cracking and rut) due to new technology in pavement and tyre design is observed. Then, a more realistic tyre–pavement contact modelling has to be considered. In order to go towards more realistic tyre–pavement contact, finite element method consumes a lot of calculation time; so in this article, we propose a semi-analytical model developed for the calculation of ball bearings. The contact is considered as a Hertzian contact, and the surface of the road is flat and that the tyre is smooth. Calculation of contact pressure is compared and validated with experimental data using a device called Stress-In-Motion, which allows contact pressure measurement. Different parameters are assumed in the modelling such as tyre pressure, axle loading and tyre shape. We have now a successful tool to analyse the tyre–pavement contact, more accurate and realistic and especially faster than those used at the moment.

A model for equivalent axle load factors

Most design methods for road pavements require the design traffic, based on the transformation of the traffic spectrum, to be calculated into a number of equivalent passages of a standard axle using the equivalent axle load factors (EALFs). In general, these factors only consider the type of axle (i.e. single, tandem or tridem), but they do not consider the type of wheel on the axles, i.e. single or dual wheel. The type of wheel has an important influence on the calculation of the design traffic. The existing design methods assume that the EALFs are valid for all pavement structures and do not consider the thickness and stiffness of the pavement layers. This paper presents the results of the development of a model for the calculation of the EALFs considering the type of axle, the type of wheel and the constitution of the pavement. The model was developed based on the tensile strain at the bottom of the asphalt layer that is responsible for bottom-up cracking in asphalt pavement, which is the most widely considered distress mode for flexible road pavements. The work developed in this study also presents the influence of the type of wheel (single and dual) on pavement performance. The results of this work allowed the conclusion that the EALFs for single wheels are approximately 10 times greater than those for a dual wheel. This work also proposes average values for the EALFs. An artificial neural network was developed to calculate the EALFs.

Primary causes of cracking of asphalt pavement in North Carolina: field study

Presently, North Carolina is experiencing higher than anticipated rates of fatigue cracking. Field materials and pavement conditions are investigated to find the primary causes of cracking in North Carolina. This study finds that pavements with high asphalt content tend to be in good condition based on the top-down cracking (TDC) observed in the top layer, and pavements with low air void contents tend also to be in good condition. In addition, fine-graded mixes tend to yield a better pavement condition than coarse-graded mixes. The structure-related comprehensive analysis suggests that debonding is one of the major causes of a poor pavement condition. Road widening is another cause of cracking that can develop into fatigue cracking. Regions in which bottom-up cracking is observed tend to have larger bending strains at the bottom of the asphalt layer than regions where TDC is observed.

Sensitivity of the mechanistic-empirical pavement design guide to traffic inputs: a space-filling approach

The mechanistic-empirical pavement design guide (MEPDG) includes a software tool to perform the large number of calculations involved in the design of pavement structures. Large number of input parameters from traffic, materials, and climate are required to use MEPDG. Data collection to accomplish calibration and implementation of MEPDG may be done more efficiently focusing on the most sensitive parameters. This research studies the sensitivity of MEPDG to traffic input parameters on two flexible pavements using the techniques for sensitivity of complex computer codes with Latin hypercube sampling, rank transformation, standardised regression coefficients, and Gaussian stochastic models. The results showed that roughness, rutting, and bottom-up cracking as predicted by MEPDG are sensitive to variations of two-way average annual daily traffic (AADT), traffic growth, dual tyre spacing, percent trucks in the design direction, percent of heavy vehicles, traffic wander, and percent of trucks in design lane. Variations of operational speed, average axle width, and mean wheel location showed no significant effects on the same MEPDG outputs. The results of the sensitivity were used to define recommended hierarchical levels for traffic input.

Fatigue life prediction of an asphalt pavement subjected to multiple axle loadings with viscoelastic FEM

A model for determining bituminous pavement degradation due to multiple-axle moving traffic loads is presented. The developmental work has been based on a viscoelastic finite element method (FEM) dedicated to bituminous pavements subjected to multiple-axle traffic loads, through reliance on the following variables: pavement structure, load speed, load configuration, and bituminous material temperature. The Load Equivalent Factor (LEF) relative to various axle configurations is approximately proportional to the number of axles within a given axle group. For flexible pavements, the fatigue life predictions using the Homsi model exceed the other two models (i.e. the French standard fatigue law, developed by LCPC–SETRA [7], and the interface fatigue law developed by Diakhaté et al. [31]). Results obtained indicate that the design concept based on bending fatigue in bituminous layers is insufficient to realistically predict the degradation of a bituminous pavement structure. The phenomenon of shear at bituminous interfaces must also be taken into account, as revealed by simulation results for the degradation of two pavement structures from the French design code [8] (i.e. flexible pavement and thick asphalt pavement). For the thick asphalt pavement structure, the main damage mode is of the “bottom-up cracking” type, whereas “interface cracking” damage is typically displayed for flexible pavements. Even before determining life expectancy by means of fatigue models, the LEF clearly indicates in every pavement structure the spot where degradation was most severe (at either the interface or bottom of asphalt layers).

Combined Effect of Three-Dimensional Contact Load and Thermal Gradients on the Cracking Performance of Heavy-Duty Asphalt Pavements

Top-down cracking is well recognized as a major distress for heavy-duty asphalt pavements, yet the mechanism is not fully understood. Studies have indicated that nonuniform tire–pavement contact stress may play a significant role in the initiation of top-down cracking, and thermal stress caused by the temperature gradient may contribute to the initiation and development of top-down cracking. These relevant factors were evaluated with a sophisticated three-dimensional viscoelastic finite element model that was developed for studying the cracking performance of heavy-duty flexible pavements under combined three-dimensional tire–pavement contact stress and thermal stress induced by the temperature gradient. It was found that the stress distribution in asphalt concrete layers continuously changed as the pavement was loaded because of the rheological behavior of the asphalt concrete. Significant maximum principal tensile stresses are present at the tire edge at the asphalt concrete surface, which may result in an accumulation of dissipated creep strain energy over time (or with a cumulative number of loads) and may eventually lead to crack initiation. Additionally, near-surface pavement responses associated with top-down cracking under three-dimensional tire–pavement contact pressure are found to be more severe than those from uniform contact pressure. In a comparison with pavements without thermal effect, the negative temperature gradient increases the potential for top-down cracking, and the positive temperature gradient increases the potential for bottom-up cracking. The study provides insight into crack mechanisms, which may improve current heavy-duty asphalt pavement design procedures.

Investigation into Top-Down Cracking of Asphalt Pavements in North Carolina

Top-down cracking has become a commonly reported cracking mechanism in asphalt pavements worldwide. In top-down cracking, a crack initiates at the surface of an asphalt concrete pavement and propagates to the bottom of the asphalt pavement layer. Because the location and governing state of stress for top-down cracking are different from those for bottom-up cracking, analysis of pavement cracking performance must be robust enough to account for the complex mechanisms that are involved in top-down cracking. This study applies the viscoelastic continuum damage finite element model to the evaluation of two pavement sections in North Carolina where top-down cracking has been identified. Small specimen geometries are used to perform simplified viscoelastic continuum damage testing on individual asphalt layers obtained from field cores. This testing model and the Fourier finite element program are used together for pavement cracking performance simulations, and the joint model uses the structure and layer material properties obtained from the two study pavements. The simulation results clearly support the propensity of these pavements to exhibit top-down cracking. In addition to laboratory testing and analysis, the deflection-based method suggested by Uhlmeyer et al. is applied to the data obtained from pavement sections with known crack initiation locations, and that method's validity is investigated. The Uhlmeyer method uses the AREA parameter, which is determined from falling weight deflectometer deflections, and pavement thicknesses. The analysis results show a clear difference in the AREA versus pavement thickness relationship between the pavement sections with top-down cracking and those with full-depth cracking.

Analysis of the Long-Term Structural Performance of Flexible Pavements Using Full-Scale Accelerated Pavement Tests

This paper presents the results from an accelerated pavement testing project carried out at the test facility of the Federal Highway Research Institute of Germany [1]. The main objective consisted in analysing and comparing the structural performance of flexible road pavements over service life. For this purpose two full-scale flexible road pavements were subjected to continued loading while capturing the structural condition. The test sections were uniformly constructed with an asphalt base course on a gravel base laid on a sub-base but differed in the thickness of the asphalt layers and the use of an asphalt binder course. During testing the surface deflection was measured and the transverse evenness was monitored. Strain gauges were used to detect changes in the structural behaviour under loading. Samples were taken from one pavement after loading to analyse structural cracking. The measurement results show a significant decrease of the structural strength of both pavements, accompanied by bottom-up cracking in the asphalt road base in one test section. In case of the test section with the thinner asphalt pavement the development of the bearing capacity and cracking in the asphalt base indicate a structural failure. The loading of the stronger pavement had to be interrupted early due to technical reasons. The comparison of the data from both sections shows significant differences in view of the total decrease of structural strength as well as the rate of deterioration.