Analysis and Performance of a Braced Cut in Sand with Large Deformations

The paper is divided into two parts. The first part deals with the systematic program of measurements undertaken on an open braced cut in dense sand at the Greenway Pollution Control Centre in London, Ontario. In the second part, the experimental data are analyzed and a new solution is presented based on Dubrova's analysis, which related qualitatively and quantitatively the active earth pressure distribution to the mode of deformation of a retaining structure.The roughly L-shaped excavation measured 68 × 42 ft (20.7 × 12.8 m) for the longest leg, the other leg was 30 × 23 ft (9.1 × 7.0 m). The temporary bracing system consisted of interlocking steel sheet piles (Larssen IIIN), and wales and struts from wide-flanged steel sections. The maximum depth of the cut was 50 ft (15.2 m) below ground elevation of 722 ft (220.1 m). The soil consisted of fine uniform dense sand having a relative density varying from medium to very dense. The natural water level was approximately 20 ft (6.1 m) below the ground surface prior to construction.The instrumentation program was carried out during the 6-month construction period (January–June 1964) and consisted of measuring: (1) The strut loads with a mechanical strain indicator (Whitmore gauge) over 8 in. (20.3 cm) gauge lengths, (2) The deformation of the north wall in a horizontal and a vertical plane, (3) The water levels and water pressures from borehole and standpipe observations, and (4) The active and passive earth pressures over the cut with 'Geonor vibrating-wire pressure transducers mounted flush on two adjacent sheet piles of the north wall.Field and laboratory tests supplied the necessary soil data.Comprehensive measurements of this kind in deep cuts in sand, prior to this London investigation, had only been made in Berlin, Munich, and New York. But at London, for the first time the actual distribution of earth pressures in sand were measured on a full-scale braced wall.The analysis of the experimental data showed that the earth pressure distribution can be approximated by the extended Dubrova’s solution. The agreement between the total active earth pressure obtained from the pressure cells and the corresponding Coulomb values varied from excellent (upper bound) to good (lower bound).An experimental relationship between the horizontal soil strain and the variation of K-values over the depth of the cut was established.The different theories for predicting Ko-values do not seem to apply to over consolidated dense sand deposits. The experimental Ko-values, rather, agree with other published experimental values for similar soils.The strut load readings were somewhat erratic, not necessarily corresponding to the excavation progress. The total strut loads were lower than the corresponding forces from the earth pressure cells or the corresponding Coulomb values.