T-bar and Ball penetration tests (TPT and BPT, respectively) were carried out at three sites consisting of soft clays. These penetration tests have several advantages over the conventional cone penetration test (CPT): much larger project area and no effects of overburden pressure for obtaining soil parameters such as undrained shear strength. Therefore, TPT and BPT are considered to be suitable in-situ tests for very soft ground. In this paper validity of the assumed mechanism of these cones are examined, using ball cones with different diameters of the ball. In addition, the cone factors for CPT, TPT and BPT are compared based on undrained shear strengths (s_u) measured from field vane and direct shear tests. Cyclic BPT was carried out to measure the sensitivity of soil layers. It is revealed that reduction in the tip resistance with the increase in the number of cyclic penetration can be correlated with the sensitivity measured by the field vane tests. Finally, the possibility of estimating the in-situ effective overburden pressure (σ′_vo) is discussed for an artificial island filled by dredged clayey soil, where consolidation due to self weight of the filling is still continuing.