Establishing the Interlayer Structural Response for Unbonded Concrete Overlays of Existing Concrete Pavements

Abstract

An improved mechanistic empirical design procedure for unbonded concrete overlays of existing concrete pavements (UBOLs) should account for the effect of the interlayer on the structural response of the pavement. One approach is to use the Totsky model to characterize the interlayer. The Totsky model treats the interlayer as a bed of springs between two plates and is currently incorporated into the rigid pavement finite element software ISLAB. A difficulty encountered in implementing this model is that there are currently no guidelines as to what the interlayer k-value should be for different types of interlayers. The interlayer can be constructed of new or aged asphalt (open or dense graded) or a nonwoven geotextile fabric. To establish the k-values that accurately characterize each of these materials, an ISLAB model of a laboratory test was created so the k-values could be established by matching the measured and calculated difference between the deflections in the overlay and existing pavement. To supplement the use of the laboratory data in establishing the Totsky interlayer k-value, an analysis was carried out using falling weight deflectometer (FWD) data from UBOLs at the Minnesota Road Research Facility (MnROAD). Analyses were then performed to determine if the difference between k-values for different interlayer materials are statistically significant, and if the results from the laboratory analysis match those obtained from the MnROAD field data. The Totsky k-value recommended for use when modeling the response of an UBOL with an asphalt interlayer is 3500 psi/in and 425 psi/in for a fabric interlayer.

Contribution of MnROAD Research to Improvements in Concrete Pavement Technology from 1994–2019

“…With the growing interest in constructing thinner unbonded concrete overlays, several test sections have confirmed that there are limits of panel size versus slab thickness. The transition from asphalt-based interlayers to non-woven geotextile fabric interlayers throughout the U.S. has been paralleled in the design of the MnROAD UBOL test sections ( 46 ).…”

“…With a minimum collection frequency of four times per year, MnROAD test sections accumulate more falling weight deflectometer (FWD) data than any other concrete pavement in the U.S. This testing frequency allows for the characterization of seasonal effects on FWD pavement responses that has been useful to the development of modern mechanical-empirical designs ( 44 , 46 ). MnROAD has also contributed toward the next generation of pavement condition devices, such as the rolling weight deflection vehicles and has hosted multiple FHWA calibration events, in which pavement management vehicles can be evaluated equally at one facility on a variety of pavement sections and surface textures ( 47 ).…”

Artificial Neural Networks for Predicting the Response of Unbonded Concrete Overlays in a Faulting Prediction Model

“…and 425 psi/in. for a fabric interlayer (8). The next step in developing the FEM model was validation using the established Totsky k-values.…”

Development of Artificial Neural Networks for Predicting the Response of Bonded Concrete Overlays of Asphalt for Use in a Faulting Prediction Model

“…These networks are separated into loaded or unloaded slab, axle type, joint depth, and temperature. Because of symmetry of the temperature loading condition, only one ANN is developed for both the loaded and unloaded sides of the joint ( 25 ). Each of the ANNs with each of their predictors are shown in Equations 3–5.…”