With much of the Netherlands’ concrete infrastructure approaching or surpassing its design life in the coming years, reliable and efficient assessment tools are essential to ensure safety and manage maintenance costs effectively. This thesis investigates the effectiveness and reliability of non-destructive testing (NDT) methods for assessing the material condition of aging reinforced concrete viaducts, with a primary case study on the Ardeweg viaduct.
This study is a direct follow-up to Gert Wilgenburg’s 2024 report on the Sluinerweg, which is another viaduct in the ”Liggerkoppen Project”. The end product of his research was a practical methodology for large-scale NDT and DT inspection. This present research aims to further develop this practical methodology by performing another large-scale inspection on the Ardeweg viaduct using various NDT methods, including: Ground Penetrating Radar (GPR), Rebound Hammer (RH), Ultrasonic Pulse Velocity (UPV), Half-Cell Potential (HCP), resistivity, and corrosion current density. The methodology combined the NDT measurements with destructive validation through core compressive strength testing, chloride profiling, carbonation depth analysis, and thin-section microscopy.
Furthermore, two types of rebound hammers (Q- and R-type) were studied to evaluate their inter changeability and variability. Particular attention was given to the effect of surface coatings. SonReb regression models were developed using RH, UPV, and destructive strength data to evaluate their accuracy for strength estimation. The Q hammer is preferred over the older R-type hammer due to its improved accuracy and user experience. The Kristal-Cement-Graniet (KCG) coating applied to parts of the viaduct has a significant effect on the RH results, increasing the variability noticeably and decreasing the mean rebound values obtained. Furthermore, results show that RH testing correlates strongly with compressive strength (2 > 0.9), while UPV ( 2 = 0.51) offers complementary value with lower sensitivity to surface conditions, like carbonation. The SonReb method, while slightly improving accuracy, showed diminishing returns relative to rebound-only models. The implementation of a carbonation correction factor for rebound values shows promise in improving the overall accuracy of the strength estimation models in both RH and SonReb models.
Corrosion risk classification from electrochemical methods aligned well with destructive chloride pro files and visual inspection, particularly in areas identified as high-risk. Half-cell potential and resistivity methods proved consistent and reliable under most field conditions, with HCP showing more consistent readings between devices and locations. The Gecor-10 and Profometer both showed a strong correlation with each other, indicating interchangeability between potential results. The Gecor-10 A sensor showed inconsistencies and more variability in both resistivity and half-cell potential measurements when compared to the other devices, raising questions about the accuracy of its calculation. The use of both 30s and 100s polarization times during inspection could provide more insight into the true value. The chloride profiles made with the RCT method in the 2019 report show a very promising correlation with the more accurate lab titrations when they are corrected for cement content. Its usefulness should be studied further as it could save time and costs.
This study concludes that NDT methods can serve as reliable tools for material condition assessment of aging concrete. Both strength- and corrosion-focused methods showed strong correlation with their destructive counterpart. Valuable practical knowledge has been obtained from the Ardeweg research, which could help shape future maintenance strategies for similar concrete structures.