EFFECT OF SURFACE-STRAIN SAMPLING INTERVAL ON THE RELIABILITY OF PRETENSIONED CONCRETE RAILROAD TIE TRANSFER LENGTH MEASUREMENTS

摘要

The feasibility of fully-automated in-plant transfer length measurement for prestressed concrete railroad cross-ties has been suggested from recent progress in the development of robust optical surface strain measurement. The non-contact Laser Speckle Imaging (LSI) technique provides rapid and accurate surface strain profile data—a key requirement if continuous monitoring of cross-tie transfer length is to be achieved in a production plant setting as a quality control parameter. Using an automated LSI system, strain data was taken at an unprecedented resolution of 0.5 inch increments during recent extensive in-plant crosstie testing. This provided a large data base for investigation of the minimum sampling interval required for accurate transfer length measurement. Using high-resolution in-plant strain data from a sampling of crossties,, the effect of sampling interval on transfer length assessment is evaluated and compared with previous theoretical analysis of transfer length measurement uncertainty. Statistical analysis of this real in-plant crosstie data shows that the influence of sampling interval on crosstie transfer length uncertainty agrees well with analysis developed earlier for constant cross-section prismatic members. Furthermore, these results indicate that it may be possible to rapidly and accurately assess transfer length with only a few discrete measurements of surface strain along the crosstie.