Interpretation of the lime column penetration test

Abstract Dry soil mix (DSM) columns can be used to reduce the settlement of embankments constructed on soft clays and to improve the stability. During construction the shear strength of the columns needs to be confirmed for compliance with technical assumptions. However, the measurement of the column shear strength can be a contentious issue. All methods of assessing the ultimate shear strength of DSM columns have limitations. These are caused by uncertainties in empirical probe factors required to convert pullout or push in force measured during the lime column penetration test to shear strength and/or testing a small proportion of the DSM column volume and determining whether it is representative of the strength of the entire column. The penetration resistance measured using the lime column test is considered to be more representative of average column shear strength than some other test types. This test can be carried out as a pullout resistance test (PORT) or a push in resistance test (PIRT). Both PORT and PIRT require empirical correlations of measured resistance to an absolute measure of shear strength, in a similar manner to the Piezocone test. In this paper, finite element techniques developed for assessment of T-bar, Ball and Piezocone penetration tests [7] , [8] , [10] are used to assess bounds for the empirical probe factor, N. To simulate the cemented DSM columns, analyses have incorporated a model for a strain softening material. Measured settlements from an embankment constructed on DSM ground improvement are then compared with finite element calculations to infer the shear strength of the columns. These inferred shear strengths are then compared with the results of PORT tests performed beneath the embankment.