Interpretation of T-bar penetrometer tests at shallow embedment and in very soft soils

TLDR

The T‑bar penetrometer, developed for centrifuge profiling of soft soils, is now a widely used offshore tool for measuring undrained strength, conventionally interpreted by converting penetration resistance to soil strength with a single bearing factor for steady flow. This study proposes a new interpretation method for T‑bar tests at shallow embedment in very soft soils, addressing a growing concern in seabed infrastructure design. The method incorporates soil buoyancy and a reduced bearing factor from shallow failure, derived theoretically and calibrated with large‑deformation finite‑element analyses. It shows that the depth at which steady deep penetration is achieved depends on the normalized soil strength \(s_u/\gamma' D\) and can reach several bar diameters, and that the revised procedure yields different strength estimates compared with the conventional approach, as demonstrated on three centrifuge samples.

Abstract

The cylindrical T-bar penetrometer was developed for profiling the undrained strength of soft soils in the centrifuge and is now a widely-used offshore site investigation tool. The conventional interpretation of the T-bar test is to convert the measured penetration resistance to soil strength using a single bearing factor associated with steady flow of soil around the bar. This paper describes a new analysis for the interpretation of T-bar penetrometer tests at shallow embedment and in soft soils, which is an increasingly significant consideration in the design of seabed infrastructure, including pipelines. The analysis captures two mechanisms that are usually neglected: (i) soil buoyancy and (ii) the reduced bearing factor arising from the shallow failure mechanism mobilized prior to the full flow of soil around the bar. The framework derives from theoretical considerations and is calibrated using large deformation finite element analyses. The depth at which the steady deep penetration condition is reached is shown to depend on the normalized soil strength, s u /γ′D, and may be up to several diameters deep. The effect of this new procedure on the inferred soil strength compared with the conventional approach is illustrated through T-bar tests in three different centrifuge samples, spanning a range of strength ratios.