Gary L Crawford

Interlaboratory Study on Measuring Coefficient of Thermal Expansion of Concrete

Coefficient of thermal expansion (CTE) is one of the sensitive inputs in the new Mechanistic–Empirical Pavement Design Guide (MEPDG). The most widely used test method to measure CTE of concrete is outlined in AASHTO TP60-00, Coefficient of Thermal Expansion of Hydraulic Cement Concrete. Several state highway agency materials laboratories and university research centers have custom-built manually operated or automated CTE measuring devices based on recommendations from AASHTO TP60. Automated CTE devices are also commercially available. With many states in the process of implementing the MEPDG, it is important that the CTE measurements from the various devices provide accurate and comparable results. Results from a nationwide CTE interlaboratory study conducted by FHWA are presented. Eleven custom-built and seven commercially purchased CTE units were used in the study. The 18 CTE devices were divided into four groups, with each group testing one 410 stainless steel (SS) specimen and two concrete specimens with low and high CTE. Statistical analysis was performed to determine overall variability of the CTE measurements from the various devices. Additionally, variability was determined between custom-built and commercially available CTE units and between two CTE measuring methodologies (AASHTO TP60 and Texas test method). Accuracy of the CTE measurements from the interlaboratory study was determined from the CTE value of a 410 SS specimen measured according to ASTM E228-06, Standard Test Method for Linear Thermal Expansion of Solid Materials with a Push-Rod Dilatometer.

Nanotechnology: New Tools to Address Old Problems

This paper outlines cement and concrete properties that challenge engineers and discusses the benefits that could be derived from changes in the smallest structure of cementitious and other concret...This paper outlines cement and concrete properties that challenge engineers and discusses the benefits that could be derived from changes in the smallest structure of cementitious and other concrete materials. Concrete is the most widely used building material in the world. Roman concrete structures still exist today. Even with concretes versatility and durability, certain properties continue to pose challenges. The ability to modify essential molecular building blocks provides the potential to make great strides in reducing or eliminating numerous mechanisms that can compromise the life of concrete. This paper discusses current durability challenges specific to transportation structures—both bridges and pavements—and the role that nanotechnology can play in addressing these issues. In structural concrete, applications could include improved tensile strength, increased ductility, reduced permeability, and reduced shrinkage. These enhancements could significantly reduce maintenance costs and greatly extend the life of most structures. Reduced shrinkage, modification of the hydration process, minimized thermal movement, reduced permeability, and improved workability would greatly extend pavement life. Nanotechnology could play a key role in environmental stewardship through significant reduction in the carbon footprint, as well as by making the cement production process more efficient. Properties that adversely affect the construction process are also feasible applications of this technology. Other modifications of the properties of concrete that could greatly increase concrete life include the following: decreasing volume change, improved mechanical performance, greater freeze–thaw resistance, reduced water migration, improved air system stability, improved properties of marginal-quality aggregates, and improved curing methods.

New AASHTO T336-09 Coefficient of Thermal Expansion Test Method: How Will It Affect You?

Although many papers were published during the past decade on the coefficient of thermal expansion (CTE) and its impact on concrete pavement design, an error was recently discovered in the AASHTO TP60-00 about the calibration of the testing equipment and, consequently, determination of the concrete CTE. The new AASHTO T336-09, even though based on the TP60-00, rectifies this calibration issue. This paper presents differences between the two test methods and implications for the Long-Term Pavement Performance database and for the Mechanistic–Empirical Pavement Design Guide and for its implementation by state departments of transportation. Recommendations are provided for improvements to the AASHTO T336-09 test method.

Ruggedness Study on the AASHTO T 336 Coefficient of Thermal Expansion of Concrete Test Method

A ruggedness study on the AASHTO T 336 coefficient of thermal expansion of concrete test method was performed to evaluate the factors most likely to affect the test results. Seven factors were evaluated: time at temperature extremes, water level, position of the linear variable differential transformer, number of segments, saturation criterion, specimen length, and temperature of the first segment. Two concrete mixtures were used in this study, four laboratories participated, and five commercially made coefficient of thermal expansion devices from two manufacturers were used. On the basis of the results obtained, saturation criterion was found to be the most significant factor. The other factors were found not to have a significant impact on the test results, have already been addressed in the most current version of the test method, or, in the authors’ opinion, do not warrant being addressed.

Interlaboratory Study and Precision Statement for AASHTO T 336 Test Method

With the recent release of AASHTOWare Pavement ME Design software, there will be a greater emphasis on measuring the coefficient of thermal expansion (CTE) of concrete because of its significance on predicted distress and design life. The most widely used test method to measure the CTE of concrete is the AASHTO T 336 test method. Data are presented from an interlaboratory (round robin) study conducted by FHWA; this study included 20 CTE units and laboratories representing FHWA, state highway agencies, universities, commercial testing laboratories, and the concrete paving industry. As part of the study, each laboratory tested nine concrete specimens (three concrete mixtures 3 three specimens per mixture) that spanned the typical range of CTE values for concrete. On the basis of this interlaboratory study for the AASHTO T 336 test method, the within-laboratory single operator standard deviation was found to be 0.12 mstrain/°C and the between-laboratory standard deviation was found to be 0.28 mstrain/°C. Data from the study also showed that there was no statistically significant difference between custom-built CTE devices versus commercially available testing devices. This demonstrates the ruggedness of the current test method.

Evaluation of the Specimen Saturation Criterion for the AASHTO T336 Test Method

Realizing the importance of coefficient of thermal expansion (CTE) in concrete pavement design, the Federal Highway Administration conducted a ruggedness study for the AASHTO T336 test method in 2012. Of the seven variables that were evaluated as part of the ruggedness study, specimen saturation criterion was found to be one of the significant variables that warranted further investigation. This paper documents a follow-up study performed to specifically evaluate the effect of specimen saturation criterion on the measurement of CTE using the AASHTO T336 test method. CTE tests were conducted on multiple specimens from five concrete mixtures (two field and three laboratory prepared) at different levels of saturation in water: T336 criterion, 4 days, 7 days, 14 days, 28 days, and vacuum saturation. Data analysis from this study indicated that there is no statistical difference in CTE measurement after 28 days of water saturation versus T336 criterion, 4 days, 7 days, 14 days, and vacuum saturation. Based on this limited study, it appears that the current saturation criterion outlined in AASHTO T336 is adequate.

COMPARISON OF PREDICTED DISTRESSES BETWEEN DIFFERENT INPUT LEVELS USING M-E PDG FOR CRC PAVEMENT

The Mechanistic Empirical Pavement Design Guide (M-E PDG) provides a number of new approaches for characterizing materials to be used in 21st Century pavement design. The M-E PDG software uses numerical models to analyze traffic, climate, subgrade and laboratory measurements of material properties to predict the performance of various pavement designs over their entire service life. A key element of the mechanistic design approach is the prediction of the response of the pavement materials, and thus of the pavement itself. The M-E PDG provides three hierarchical input levels: Level 1 is site/project specific with actual tests resulting in higher accuracy, Level 2 from less than optimal testing or by correlations and Level 3 from the agency database or user selected default values. Since obtaining Level 1 material inputs require considerable testing effort and cost, Level 1 data for M-E PDG analysis may not be possible all the times. In such situations, Level 2 or 3 inputs can still be used to perform M-E PDG analysis and predict design life of pavement. The objective of this study is to compare M-E PDG predicted distresses of different input level data from a recently constructed Continuously Reinforced Concrete Pavement (CRCP) project at I-64 Battlefield blvd, Chesapeake, Virginia. The original design of this pavement was done using the 93 AAHSTO design method several years before the construction. This paper also documents the efforts associated with the laboratory and field procedures used for level 1 input characterization. The FHWA Mobile Concrete Laboratory (MCL) had performed all the concrete materials characterization for level 1 input on this project. M-E PDG software (version 1.1) is used to obtain the predicted distresses for the pavement structure using the same thickness obtained by AASHTO method. This paper also discusses the difference in predicted distresses when level 1 and level 3 traffic data was used. Results shows that there were significant differences in predicted distresses between Level 1 and Level 3 concrete properties inputs for M-E PDG analysis depending upon the type of Level 3 input used.