Number of Replicate Beams Required for a Valid Test of Asphalt Mixtures Using the Bending Beam Rheometer Based on AASHTO TP125
Abu Sufian Mohammad Asib, University of UtahShow Abstract
Pedro Romero, University of Utah
Faramarz Safazadeh, University of Utah
Developing any test, standard, or specification requires technical validity while keeping the costs as optimum as possible. AASHTO TP125 recommends Bending Beam Rheometer (BBR) as a simple, fast, relatively inexpensive, and repeatable method of testing low-temperature properties of asphalt concrete mixtures. However, a question on the number of replicate beams required to be a valid representation of population for asphalt mixture beam specimens remains unaddressed. This study used statistical methods to determine this number by analyzing the data of two variables- creep moduli and stress relaxation capacity (m-value); obtained from testing two different asphalt concrete mixtures. It was found that BBR data follow a normal distribution with minor skewness. Moreover, by randomizing the population and taking a different number of replicate beams to analyze coefficient of variation (CV) and sample means, it was found that eight beams represent the tests results of the BBR with a CV ranging between 0.05 and 0.11, and the maximum difference between sample and population mean is below 7%. Even though testing five beams as suggested in AASHTO TP125 showed more variability in the results, still the CV was well below 0.2, and the sample mean showed more than 10% difference from the population mean. In practice, core samples of Hot Mix Asphalt (HMA) provide a limited number of replicates; therefore, testing eight beams in both field and laboratory is feasible. Results obtained from this study suggests BBR as a dependable testing equipment to characterize the low-temperature performance of asphalt concrete mixtures.
Variability Analysis of Asphalt Mixture Beam Bending Test
Wanqiu LiuShow Abstract
Wen Lu, Dalian University of Technology
Xu Liu, Dalian University of Technology
Zheren Wang, Dalian University of Technology
The random shape and distribution of the aggregates add large inhomogeneity into the mixtures. Large material composition variability will increase the risk of damages, such as cracking and rutting. This paper introduced a digital image processing based HMA mixture material composition variability analysis method as a new material performance evaluation supplementary tool. The low temperature performances of six types of asphalt pavement materials have been evaluated through beam bending tests. Through comparing with the beam bending lab test data, it has shown that material composition variability has direct influence on the variability of the strength of the HMA beam bending test specimens, and proved the effectiveness of the proposed variability analysis method.
Effect of Loading Waveform Pattern and Rest Period on Fatigue Life of Asphalt Concrete Using Viscoelastic Continuum Damage Model
Waleed Zeiada, University of SharjahShow Abstract
Padmini Gudipudi, Flatiron Constructors, Inc.
Mena Souliman, University of Texas, Tyler
Fatigue cracking is one of the most critical distresses in asphalt pavements and is due to actions of repetitive traffic loading over time. The fatigue life of asphalt concrete is often estimated from laboratory experiments where the performance depends directly on the test method, loading conditions, temperature, rest period, and aging in addition to the composition and properties of the mixture itself. The uniaxial fatigue test has become a popular method for developing constitutive models that describe the fatigue behavior of asphalt concrete mixture owing to the uniform states of stress across the specimen section. This study investigates the effect of the loading waveform (sinusoidal vs. haversine) and rest period (continuous vs. intermittent) on the laboratory fatigue life of asphalt concrete mixtures. The fatigue analysis was performed using the simplified viscoelastic continuum damage (S-VECD) approach where the damage characteristic (C-S) curves were established for all the cases, and then used to estimate the fatigue laws through simulated predictions. The proposed uniaxial fatigue test and analysis method were able to determine the fatigue life relationships of asphalt concrete mixture at different waveform and rest period conditions with a reduced testing time compared to other traditional testing and analysis methods. Overall, both rest period and waveform pattern were found to affect the laboratory fatigue life of asphalt concrete mixture. Model predictions show that pulse-rest loading yields an equivalent fatigue life to continuous loading at strain values that are approximately 4 times greater.
Sine Versus Haversine Displacement Waveform Comparison for Hot-Mix Asphalt Four-Point Bending Fatigue Testing
Angel Mateos, University of California, BerkeleyShow Abstract
Rongzong Wu, University of California, Davis
Erik Denneman, Australian Asphalt Pavement Association (AAPA)
John Harvey, University of California, Davis
An experimental study was conducted to determine the effect of haversine and sine displacement waveforms on four-point flexural fatigue test results for hot mix asphalt. Seven asphalt mixtures with different gradations, binder types, and binder contents were tested for this study. Four of the mixes were tested in California and three in Australia. The mixes were tested at different strain levels under both waveforms in displacement-controlled mode without rest periods. The comparison showed no differences between haversine and sine testing modes for six of the seven mixtures. Fatigue life, the shape of the stiffness reduction curves, initial dynamic moduli and phase angles, and Black diagrams were essentially the same for both testing modes. This similarity was attributed to the viscoelastic nature of the asphalt mix. Because of asphalt viscoelasticity, the beam at-rest position in haversine testing will rapidly move to halfway between zero and maximum displacement, and so the same stress is produced by haversine and sine displacement waveforms as soon as the peak-to-peak amplitudes are equal. For one of the seven mixes, sinusoidal testing produced considerably longer fatigue lives, although it is believed this mix is an outlier and can be ignored. Based on these results, it was concluded that a practical reason to recommend sinusoidal versus haversine mode of testing is that the latter becomes sinusoidal after a small number of load repetitions. For the same reason, the harmonization of the ASTM and AASHTO tests to require a sinusoidal mode of testing is recommended.
Conceptual Development of Nomograph to Estimate Modulus with Progressive Fatigue: An Inclusive Application of Dynamic Semicircular Bending Test
Gourab Saha, Rowan UniversityShow Abstract
Krishna Prapoorna Biligiri, Indian Institute of Technology, Tirupati
The objective of this research study was to estimate the degradation of asphalt materials properties measured by modulus based on the continuous fatigue process with the help of fracture characteristics. A total of eighteen asphalt mixtures were prepared, and dynamic semi-circular bending (SCB) test was conducted at three test temperatures. A series of parameters were determined and evaluated to understand the fracture properties of various asphalt mixtures using linear elastic fracture mechanics (LEFM) principles. In addition, the initial moduli of all the mixtures were characterized using |E*| dynamic modulus test, and fracture properties obtained from dynamic SCB test were employed to estimate the reduction coefficient (δSCB) with respect to the initial modulus. The study framework offered the flexibility to change the input variables during the calculation procedure of δSCB. Further, a nomograph was developed using graphical relationship amongst the fracture properties to estimate δSCB for quick approximation purposes. Overall, this study established the correlation between reduction in the modulus and fracture properties, and further developed the conceptual outline to formulate a nomograph that could conveniently estimate the residual modulus of asphalt mixtures with respect to specified lengths of crack growths. It is envisioned that the procedure developed in this study would be useful for comprehensive evaluation of fatigue cracking of asphalt mixtures, and can potentially serve as a helpful tool to estimate/forecast residual life of different asphalt pavements.
Semicircular Bending (SCB) Test on Lab-Produced, Plant-Produced, and Field Warm-Mix Asphalts
Amir Arshadi, AECOMShow Abstract
Rouzbeh Ghabchi, South Dakota State University
Syed Ashik Ali, University of Oklahoma
Manik Barman, University of Minnesota, Twin Cities
Musharraf Zaman, University of Oklahoma
Sesh Commuri, University of Nevada, Reno
Fatigue cracking is a major distress in asphalt pavements. Occurrence of fatigue cracking depends on many factors such as pavement structure, traffic, environment, and most importantly fatigue performance of asphalt mixes used in the construction. Application of Warm Mix Asphalt (WMA) technology and incorporation of Reclaimed Asphalt Pavement (RAP) and Recycled Asphalt Shingles (RAS) in asphalt mixes are effective and sustainable ways to preserve the environment while reducing the construction cost. Despite several advantages, concerns over the long-term performance of the green asphalt pavements have limited their use. In the present study, the cracking resistance of five different asphalt mixes including one hot mix asphalt (HMA) and four WMA mixes, were investigated using the Louisiana Semi-Circular Bend test, called SCB test in this paper. For all of the five mixes, the tests were conducted on laboratory-produced mixes, plant-produced mixes and, field cores. The results indicated that application of softer virgin binder and/or recycling agent improves the cracking resistance of a mix. Furthermore, the test results conducted on laboratory-produced and plant-produced mixes as well as the field cores were compared. Based on the outcomes of the laboratory tests, the SCB test was found to be an effective tool for screening of mixes for cracking resistance in the mix design process. The findings of this study are expected to add to the knowledge base in the area of the long-term performance of green asphalt pavements and facilitate their use, nationwide.
Fatigue Susceptibility of Asphalt Paving Mixtures Incorporating Random Rest Period and Environmental Factors
Ahmed Aljubory, No OrganizationShow Abstract
Gordon Airey, University of Nottingham
James Grenfell, University of Nottingham
The prediction of fatigue damage of bituminous materials has a great importance in the design of asphalt pavements. It is supposed to be accelerated by the combined action of environmental factors and repeated loading. Therefore, it is important to incorporate the environmental conditions in that predicting to ensure a proper design period of the in service pavement structure. This study aims to explore more about the effects of LTOA and moisture damage incorporating random rest periods on the fatigue susceptibility of asphalt concrete mixtures. The effect of LTOA and moisture was simulated using British Board of Agrément (BBA) protocols and evaluated using ITFT (Indirect Tensile Fatigue Test) in strain controlled mode. A 14mm Dense Bitumen Macadam (DBM) mixture was manufactured using a 40/60 penetration asphalt binder and limestone aggregate. The results of ITFT at both testing temperatures 10°C and 20°C show a decrease of control and LTOA asphalt mixtures fatigue life after moisture conditioning. Moreover, moisture reduces the stiffness recovery of these mixtures compared to those under dry conditions. On the other hand, there was an increase in the fatigue life of 14mm DBM mixtures when these mixtures stiffen due to LTOA.
Evaluation of the Fatigue Performance of Fine Aggregate Matrices Prepared with Reclaimed Asphalt Pavements and Shale Oil Residue
Andrise Klug, University of Sao PauloShow Abstract
Adalberto Faxina, Universidade de São Paulo
Amit Bhasin, University of Texas, Austin
Andressa Ng, Universidade de São Paulo
Higher proportions of reclaimed asphalt pavements (RAP) have been added to asphalt mixtures, aiming to reduce the use of virgin materials and construction costs. One limitation of such technique is the increase of the mixture stiffness, caused by the aged binder. Stiffer mixtures are more prone to cracking and rejuvenating agents have been used to reduce the stiffness of the aged binder. The cracking process begins as micro cracks at discontinuities of the fine portion of the asphalt mixture, and studies with the fine aggregate matrix (FAM) have been developed, based on such interpretation of the fatigue phenomenon. Such technique was used on this work in order to evaluate the incorporation of RAP and rejuvenating agent on the fatigue performance of new asphalt mixtures. One source of RAP was used at three percentages (0, 20 and 100%) and a shale-oil residue was used as rejuvenating agent at three percentages (0, 50 and 100%). The addition of RAP increases the stiffness and the damage evolution rates of the FAMs, resulting in lower fatigue lives, as compared to the reference FAM. The FAM prepared with new asphalt presented fatigue performance greater than the FAMs prepared with asphalt/agent ratios of 50/50 or 0/100. The use of only a new binder of PG 64 is the best option to correct the binder content of FAMs produced with 20% of RAP. The unexpected bad performance of the shale oil residue is probably due to the low diffusion rate of the material into the aged binder.
Evaluation of the Fatigue Damage Behavior of Fine Aggregate Matrices Prepared with Modified Binders Using the VECD Theory
Andressa Ng, Universidade de São PauloShow Abstract
Adalberto Faxina, Universidade de São Paulo
Amit Bhasin, University of Texas, Austin
Andrise Klug, University of Sao Paulo
Good agreement has been observed between the properties of fine aggregate matrices (FAMs) and HMA mixtures in studies related to moisture, fatigue and rutting performance. Because of that, researchers have been studying FAMs as a tool to assess the behavior of HMA mixtures under different distress mechanisms. Investigations on the effects of modified binders and binder aging on the fatigue performance of asphalt mixtures are relevant issues, once that modified binders can enhance the fatigue behavior, and binder aging, in turn, is capable of hardening the asphalt binder, with negative effects on the fatigue performance. In order to address such issues, this paper has the objective of evaluating the effect of modifiers (SBS copolymer and polyphosphoric acid – PPA), and aging level (short-term and long-term at 30 days) on the fatigue life of fine aggregate matrices. The FAMs samples were tested in the DSR under stress control at 25°C and the results were treated by means of the theory of the viscoelastic continuum damage (VECD). The overall results pointed out that the modified binders are capable of enhancing the fatigue lives of the FAMs. An exception to this rule was observed for the FAMs aged in long-term at 30 days at high strains (around 10 %), where the FAM prepared with the neat binder presents higher fatigue lives. The FAMs produced with the AC+PPA and the AC+SBS presented the same position in the rank order, what means that these materials present similar fatigue lives, regardless of the aging level.
Prediction of Fatigue Cracking in Flexible and Semirigid Asphalt Pavement Sections
Iuri Bessa, Universidade de São PauloShow Abstract
Kamilla Vasconcelos, University of Sao Paulo
Veronica Castelo Branco, Universidade Federal do Ceara
Luis Nascimento, Petrobras
Liedi Bernucci, Universidade de São Paulo
The prediction of asphalt pavements performance in relation to their main distresses has been the focus of different studies. The characterization of fatigue cracking resistance has been proposed by different researchers, by means of laboratory testing and field data evaluation. In relation to fatigue cracking characterization, there is no consensus on which type of test should be done, the failure criterion to be considered, the test conditions (loading frequency, stress levels and temperature) and the specimens’ geometry. In terms of laboratory characterization of asphalt mixes, the most common tests are the indirect tensile test (ITT), the four point bending beam test (4PBBT), and the tension-compression test. The results obtained from these tests have been used in fatigue cracking prediction models. The present research performed the characterization of one hot mix asphalt (HMA) in terms of fatigue cracking resistance. Then, the results were used in different performance models, including one newly proposed mechanistic-empirical method for Brazilian asphalt pavement roads design that considers the fundamental properties of the asphalt mixes tested. The results indicate that the conventional fatigue prediction models provided by the Asphalt Institute (AI) and Shell Oil tend to underestimate the fatigue life of the asphalt pavement structured analyzed. The main reason could be that these models do not consider relevant inputs, such as base layer stiffness, climate conditions, and the viscoelastic characteristics of the asphalt mix. The proposed Brazilian design method resulted in predicted values of cracked area that are very similar to the values observed in the field.
Evaluation of Fatigue Characteristics of In-Service Cold Recycling Mixture with Asphalt Emulsion and HMA Mixture
Huailei Cheng, Tongji UniversityShow Abstract
Liping Liu, Tongji University
Lijun Sun, Tongji University
Cold central-plant recycling (CCPR) mixture with asphalt emulsion has been successfully used in heavy-duty pavement rehabilitation for 11 years, and now more frequently for its effective environmental protection and obvious cost advantage in China. Therefore, the fatigue performance of CCRP attracts the engineer’s and researcher’s attention. This study aimed at investigating the fatigue characteristics of CCPR mixtures that cored from in-service pavement with indirect tensile fatigue test (ITFT). At the same time, the fatigue characteristics of hot mix asphalt (HMA) mixture cored at same location were also evaluated for comparison. The fatigue test results demonstrated that actual traffic loads led to reduction in the initial stiffness modulus and laboratory fatigue lives for both CCPR and HMA mixtures, and that both in-service mixtures have the same ITFT-based fatigue damage characteristics. With increasing load repetitions, the stiffness modulus of in-service mixture declined rapidly to 45% of the initial stiffness, then entered stable fatigue expansion stage, and when the stiffness modulus declined to 20% of the initial stiffness, the mixture cracked. Moreover, it was found that the indirect tensile strength could not reflect the fatigue performance for in-service mixture, but the initial stiffness modulus could be used as an indicator to evaluate mixture’s fatigue performance.
Optimization of the Laboratory Fabrication of Small Specimens for Asphalt Mixture Performance Testing
Sonja Pape, North Carolina State UniversityShow Abstract
Kangjin Lee, North Carolina State University
Cassie Castorena, North Carolina State University
Y. Richard Kim, North Carolina State University
The use of 38-mm diameter small specimens for uniaxial dynamic modulus and cyclic fatigue asphalt mixture performance testing offers a significant opportunity to improve the efficiency of laboratory-fabricated specimen testing because multiple test specimens can be extracted per Superpave gyratory-compacted (SGC) sample. This study seeks to optimize the procedure used for the extraction of small specimens from SGC samples for dynamic modulus and cyclic fatigue tests. To this end, small cylindrical specimens were cored horizontally and vertically from SGC samples and subjected to performance testing. The dynamic modulus and fatigue test results indicate that the effects of anisotropy are minimal. However, all of the horizontally-extracted small specimens exhibited fatigue failure at the specimen ends, outside the range of the gauges; the failure was likely due to the peripheral air void gradients in the SGC samples. Therefore, the authors concluded that small specimens should be vertically-cored from SGC samples for the laboratory fabrication of small specimens. Specifically, four small specimens were cored vertically from the inner 100 mm of SGC samples where the air void content is relatively uniform. Four mixtures with different nominal maximum aggregate sizes (NMASs) were used to prepare small specimens using the proposed extraction procedure. These specimens were subjected to dynamic modulus and cyclic fatigue testing. The results demonstrate an increase in specimen-to-specimen variability with an increase in NMAS, which also is expected in large specimen testing.
Effect of I-FIT Configuration on Test Results of Asphalt Mixtures
Edoardo Barber, Mead and Hunt, Inc.Show Abstract
Hasan Ozer, University of Illinois, Urbana Champaign
Imad Al-Qadi, University of Illinois, Urbana Champaign
Zehui Zhu, University of Illinois, Urbana Champaign
Shenghua Wu, University of South Alabama
The Illinois Flexibility Index Test (I-FIT) was developed to provide a practical and reliable parameter for the evaluation of asphalt concrete (AC) fracture resistance. The flexibility index (FI) can capture the behavior of AC in terms of their ductile or brittle fracture propagation. Through the evaluation of fracture energy and the post-peak slope of the load-displacement curve, the FI is capable of detecting variations in the overall resistance of AC against crack propagation. To ensure a reliable test, potential sources of variability should be investigated as part of the test development. The differences related to equipment settings, including the loading system and fixture components, are among the potential sources of variability. This paper presents results from the machine comparison study, carried out in relation to the use of different I-FIT devices. Three different combinations of devices were compared using eight different AC mixes. A statistical analysis was implemented on the test results to evaluate if any significant differences could be highlighted among average FI values given by each machine. The tests covered eight AC mixes using three I-FIT machines; results did not show any significant difference among the devices. In particular, the loading system and the support fixture did not influence FI results.
Effect of Mix Parameters on the Semicircular Beam Fatigue Test
Xuan Chen, Pennsylvania State UniversityShow Abstract
Mansour Solaimanian, Pennsylvania State University
The fatigue of asphalt mixtures has long been a complicated problem. The Semi-Circular Beam (SCB) test has shown potential as a promising test to quantify fatigue resistance of asphalt mixtures. To determine the effect of asphalt mix composition on the SCB fatigue test results, a total of 200 SCB tests were conducted in this study using a modified SCB test procedure based on the recently developed flexibility index (FI) concept. Tests were conducted on both short and long term aged asphalt mixes. The mix components considered as test variables included binder content (BC), air void (AV), and binder stiffness. The results showed a significant reduction in fatigue resistance after long term aging. Test results also indicated that both fracture energy and flexibility index are good indicators in differentiating the effect of binder content and binder performance grade, with FI being a better indicator. Test results indicated that the FI increases with the increase of air voids. This trend is contrary to the expected trend in fatigue performance. As a result, an approach was taken to modify FI taking into account the air void of the mix, delivering reasonable results.
Improving the Reliability of Damage Characteristic Curves in the Simplified Viscoelastic Continuum Damage Model
Kangjin Lee, North Carolina State UniversityShow Abstract
Cassie Castorena, North Carolina State University
Y. Richard Kim, North Carolina State University
One of the major advantages of the cyclic fatigue test (AASHTO TP 107) is that the results can be used to calibrate the Simplified Viscoelastic Continuum Damage (S-VECD) model, which is used for mechanistic pavement performance predictions. The crux of the S-VECD model is the damage characteristic curve, which has been shown to be independent of mode of loading, loading history, and temperature. Consequently, a model can be fitted to the damage characteristic curve and used to predict the damage response for any given loading history of interest using limited test results. AASHTO TP 107 currently lacks a specific procedure for fitting a model to the damage characteristic curve and evaluating the repeatability of test replicates. In this study, a robust and practical method is proposed for fitting a power law model to the damage characteristic curve. The proposed fitting method was verified using cyclic fatigue test results of 19 mixtures sourced from the United States, Canada, and South Korea. In addition, a means to evaluate the specimen-to-specimen variability of damage characteristic curves using a shape factor is proposed. Thresholds for acceptable variability in the shape factor were derived using confidence interval analysis and verified through FlexPAVE™ pavement performance predictions. The findings of this study can be used to improve the reliability of the damage characteristic curves derived from cyclic fatigue tests for pavement performance predictions.
Characterization of Crack Growth Rate of Sulfur-extended Asphalt Using Cyclic Semi-circular Bending Test
Jun Zhang, Texas A&M UniversityShow Abstract
Maryam Sakhaeifar, Texas A&M University
Dallas Little, Texas A&M Transportation Institute
Amit Bhasin, University of Texas, Austin
Characterizing crack propagation of asphalt mixtures is helpful for optimizing mixture design and predicting cracking performance of asphalt pavements. This paper develops a new method based on the cyclic semi-circular bending test to characterize crack growth rate of asphalt mixtures. Compared with other fatigue tests, a cyclically-loaded semi-circular test has the advantages of simple specimen geometry and high efficiency of sample preparation. To accurately capture crack length for determining crack growth rate, a digital image correlation is used, and crack mouth opening displacement is measured by linear variable differential transformers mounted on the surface of the specimen. Correlations between crack length and crack mouth opening displacement are established, which are used to determine crack lengths corresponding to loading cycles over the testing process. The proposed cyclic semi-circular bending test successfully characterizes the Paris’s law coefficients of sulfur-extended asphalt mixtures. The cyclic semi-circular bending test provides substantially lower coefficients of variance in terms of cycles to fatigue failure compared with other traditional fatigue tests such as the bending beam fatigue test and the Texas Overlay Test. Test results were used to determine the impact of sulfur content on fatigue life with the conclusion that a low level of sulfur added in the case of 15% change the rheology (softener) of the asphalt to the degree that more damage is caused in a controlled-stress mode loading. However, an increase in sulfur content (30% and 45%) apparently produces a stiffer mixture that is more resistant to damage and is comparable to the control mixture.