Assessment of Suction Caisson Drained Tensile Capacity through Multiscale Physical Tests-Paper

Paper

Pdf Summary

The study evaluates the drained tensile capacity of suction caissons in dense, saturated sand using single-gravity physical tests at multiple scales. Tests were conducted on caissons with diameters of 193, 270, 400 and 1,400 mm. After suction-assisted installation, each caisson was pulled out quasi-statically under drained conditions; measured resistance was adjusted for buoyant weight and residual suction beneath the lid.<br /><br />Dimensionless comparisons showed a scale effect: normalized peak tensile resistance decreased as caisson diameter increased, with the effect becoming less pronounced at larger scales. A simple normalization described the load–displacement response reasonably well for the three largest models, but did not fully account for scale effects in the smallest model. Back-calculated interface resistance suggested that, for larger caissons with comparable geometry and soil conditions, a Ktan(δ) value of approximately 0.30–0.40 may be suitable for preliminary design.<br /><br />The study also proposed a preliminary approach to estimate initial tensile stiffness, though its scale-dependent trend requires further validation. Compared with an empirical extrapolation from the tests, API and DNV GL methods—developed for driven piles—underestimated skin friction at small scales and overestimated it for larger caissons. The method of Houlsby et al. (2005) gave the closest agreement, particularly for the extrapolated 6 m caisson.<br /><br />Overall, the results demonstrate the value of multiscale physical testing for understanding scale effects and assessing design methods. Further testing and numerical studies are needed to confirm the findings and extend their applicability.

Keywords

suction caissons

drained tensile capacity

scale effects

dense saturated sand

physical model tests

tensile resistance

interface friction

load-displacement response

initial tensile stiffness

design methods