Evaluation of Joint and Crack Load Transfer Final Report

Previous | Table of Contents | Next

The analyses reported in this document were intended to calculate LTE and backcalculate stiffnesses of cracks and joint of PCC pavements from deflection measurements, as well as joint opening from gage distance data, and recommend representative parameters for each section for inclusion in the LTPP database.  This report presents a discussion on the selection of the computational methodology and the results of the calculations of LTE and stiffness of cracks and joints for GPS, SPS, and SMP rigid pavement sections, as well as changes in joint opening for SMP sections.  The following are highlights of the study:

• Large amounts of high-quality LTE data have been collected under the LTPP program.  This information will be a valuable resource in improving the understanding of load transfer effects and improving pavement design and rehabilitation procedures.
• LTE was found to be a complex parameter that depends on many factors, including the presence of dowels, base type, subgrade, FWD load plate position (approach and leave side), testing time, and season.
• It was found that nondoweled JCP with a significant level of faulting had a much higher percentage of sections with a low level of LTE than sections that did not exhibit significant faulting (sections with high faulting had lower LTE).
• Testing time and season were found to have profound effects on LTE.  For the same joints, LTEs measured early in the morning in a cool season were found to be much lower than LTEs measured in the afternoon in warmer weather.  This effect was much greater for nondoweled JCP.
• A strong correlation was found between LTE and PCC slab temperature when higher temperature resulted in higher LTE joints.
• LTE of CRCP cracks was found to be very high (more than 98 percent of cracks had LTE greater than 80 percent).
• LTE of doweled joints was found to be much higher than LTE of nondoweled joints (more than 66 percent of doweled joints had LTE greater than 80 percent, whereas only 44 percent of nondoweled joints had LTE greater than 80 percent).
• LTEs from leave and approach side deflection testing data were found to be statistically different for more than 20 percent of JCP sections.  The difference in LTE was greater for sections with lower LTE. An attempt to explain this phenomenon by the presence of a void under the leave side of the joint was not successful. The hypothesis that this discrepancy may be caused by test sequence could not be verified under this study.
• LTE of CRCP cracks was found to have lower variability (crack to crack), load level dependency, and FWD load position dependency than LTE of JCP joints.
• LTE of doweled joints was found to have lower variability, load level dependency, and load position dependency than LTE of nondoweled joints.
• The effect of dowel diameter on LTE of doweled joints was not clearly identified.  However, this may be explained by confounding effects of other factors, such as traffic level.  In some cases, for large diameter dowels, the LTE was fairly low (less than 40 percent).  Poor consolidation of PCC around dowels may significantly reduce dowel effectiveness.
• LTE of nondoweled joints was found to be slightly lower for sections with a lean concrete base compared to sections with other base types.
• Nondoweled joints of sections with fine subgrade exhibited slightly higher LTE than those with a coarse subgrade.
• Poor correlation was found between LTE and design parameters such as PCC thickness, PCC strength, design steel content, joint spacing, and joint orientation. 
• Joint width changes daily and seasonally.  Mean daily opening and closing were found to be within plus or minus 1 mm (0.039 inch).  Mean seasonal opening and closing were found to be within plus or minus 3 mm (0.117 inch). 
• Strong correlation was found between PCC slab temperature and joint opening and closing. In addition, only a few sections showed good relationships between LTE and joint opening.
• A backcalculation procedure for joint stiffnesses was developed based on Crovetti's equation. From the results of backcalculation for the LTPP sections, typical ranges were recommended for selection of joint stiffnesses for CRCP cracks, doweled and nondoweled joints for use in finite element analysis.
• With the current procedure to measure LTE with the FWD sensor configuration, LTE may need to be adjusted slightly for slab bending occurring between the joint and FWD sensors.  An assumption of symmetrical joint behavior and full contact between the slab and the foundation was made.  This assumption, however, is most likely invalid in the field.  A more sophisticated correction factor needs to be developed.

Design, construction, and evaluation implications:

• CRCP with adequate reinforcement has very high crack LTE, which controls crack deterioration and punchouts.
• Use of dowels provides joint LTE that is higher and more consistent over time and seasons. 
• Poor LTE occurred on some doweled joints.  This may lead to premature faulting.  The cause may be a lack of consolidation around the dowel, which may need to be checked at construction.
• It is important to conduct FWD LTE testing early in the morning in cool weather to provide a realistic estimation of LTE during times when the most damage occurs.

To improve procedures for evaluation of PCC joints and cracks, the most urgent needs are to:

• Develop a mechanistic model for PCC cracks/joints capable of describing load position dependent behavior.
• Develop a better understanding of the effects of joint/crack opening and PCC slab curling on LTE.
• Develop a better understanding of the effects of permanent voids (erosion) under the joint on LTE and faulting.
• Develop a bending correction factor accounting for nonsymmetrical joint behavior, presence of a void, and PCC slab curling.
• Perform multivariate analysis to identify the effect of design features and site condition on LTE accounting for interaction between these parameters.
• Perform time history analysis of FWD deflection data for tests J4, C4, J5, and C5.
• Investigate the effects of LTE deterioration on development of base/subgrade erosion, pumping, joint faulting, and CRCP punchouts.

These additional studies will significantly improve our understanding of rigid pavement behavior and significantly contribute toward the improvement of mechanistic-empirical design procedures.

Previous | Table of Contents | Next