PREVENTION OF LONGITUDINAL CRACKING IN SURFACES AND FILLS. FINAL REPORT

Longitudinal cracking was discovered in flexible pavement constructed over high plasticity index (high-PI) clay embankments retained by reinforced soil retaining walls (RSRWs). The cracks were noted to occur at approximately the location of the interface between the clay embankment and the cohesionless backfill of the RSRW. Coring of the crack revealed the crack to be wider at the bottom than at the top suggesting that the tensile force causing the cracking was being applied at the bottom of the pavement structure. All Texas State Department of Highways and Public Transportation (SDHPT) Districts were queried to learn the extent of the longitudinal cracking problem. Only District 16, headquartered at Corpus Christi, was found to be experiencing the problem. Using soil from the same source as was used to construct embankments and RSRW backfill, four large laboratory models were constructed of the field situation. Each of the tests demonstrated that a vertical separation would occur between the clay embankment and sand backfill as a result of lateral shrinking of the high-PI clay over time due to climate. The third test showed that the vertical crack would propagate upwards through the flexible base material. The fourth test produced a crack through the flexible base but failed to produce a crack through the overlying HMAC because the HMAC adhered to the walls of the test box and separated from the caliche base. Lateral stresses imposed on the flexible base by the laterally shrinking clay embankment were found to be approximately 366 psi in the laboratory tests, well exceeding the tensile strength of normal HMAC. Thus, it was concluded that the longitudinal cracks in the District 16 pavements were the result of the clay embankments laterally drying beneath the installed pavement. Field instrumentation installed during the study was inconclusive with respect to measuring soil moisture condition changes occurring beneath the pavement because the monitoring sites were not paved until nearly the end of the study. However, the instruments showed that a change in soil suction pressures of up to 25 atm of pressure had occurred during the 1-year monitoring period as a result of changes in climate only.