Published: Jan 1989
| ||Format||Pages||Price|| |
|PDF Version (276K)||16||$25||  ADD TO CART|
|Complete Source PDF (14M)||16||$109||  ADD TO CART|
Accurate deflection readings must be obtained when using NDT devices in order to obtain correct backcalculated layer moduli. This paper provides an analysis of the sources of deflection errors and illustrates through a series of examples how random deflection errors and random pavement thickness errors affect backcalculated moduli.
The consequence of these errors is most pronounced in the backcalculated modulus of the surface course. The moduli of the base and subbase layers are also affected, and the subgrade moduli are only slightly affected by these errors. To some degree this outcome may be due to the bottom-to-top approach used by the backcalculation program MODCOMP 2.
The consequence of the variability of the backcalculated moduli on the required pavement overlay thicknesses was found to be minimal. This was due to the fact that, generally speaking, data errors which cause an overestimate of the base course modulus also cause an approximately equally significant, but opposite, underestimation of the surface course modulus. Thus, the needed overlay requirement remains nearly the same.
The paper shows how certain FWD calibration and testing procedures can be used to minimize deflection measurement errors. The accuracy of backcalculated moduli is shown to be determinable from the RMS error of the deflection fit. The examples provide an impression of the extent to which random layer thickness errors and deflection measurement errors will influence the ability to discern point-to-point variability of layer moduli for the specific pavement structure considered in this investigation. This provides a perspective from which the practical accuracy of the backcalculation approach can be judged.
backcalculation, layer thickness, deflection measurement errors, NDT, non-destructive testing of pavements, pavement engineering, elastic moduli, FWD, falling weight deflectometer
Cornell University, Ithaca, NY
Texas A&M University, College Station, TX
Dynatest Consulting, Inc., Ojai, CA
Paper ID: STP19810S