On the effectiveness of sand drains: Discussion

Abstract

Casagrande and Poulos ( 1969) have made a thorough and interesting study of the effectiveness of sand drains on the basis of published and unpublished data from a number of case records. Among the cases discussed by the authors are some fullscale tests carried out by the Swedish Geotechnical Institute at SkC Edeby (about 25 km west of Stockholm) which have been reported by Hansbo (1960). The measurements at Ski-Edeby are still going on and some additional information about Test Areas I1 and IV are presented below. Test Areas I1 and IV have a diameter of 35 m (115 ft) and are loaded by a 1.5 m (5 ft) high gravel fill, which corresponds to a surface pressure of 2.7 t/m2 (0.27 t.s.f.). Mandeldriven sand drains with a diameter of 18 cm (7 in.) and spaced 1.5 m (5 ft) apart were installed in Test Area 11. In Test Area IV no drains were used. Typical results from Test Areas I1 and IV are shown in Fig. 1. The indicated overloads in Fig. 1 refer to the middle of the two test areas. The piezometric head was measured at the middle of the test areas and at three points about 17 m (57 ft) outside the test areas at a depth of 5.0 m (16.5 ft). The compression shown in Fig. 1 relates to the middle of the test areas and the layers which initially were located between 2.5 m (8.3 ft) and 7.5 m (25 ft) below the ground surface. The pore pressure gages in Test Area I1 were located halfway between two sand drains. It should be noted that the measured pore pressures are higher than the mean values until the excess pore pressures have fully dissipated between the drains. In April 1959, when the last measurements shown in Hansbo's report were made, the measured piezometric level in Test Area N at the depth of 5 m was only about 0.7 m higher than in Test Area 11. The difference between the mean values must, however, be larger than 0.7 m because of the location of the pore pressure gauges in Test Area I1 as mentioned above. As shown in Fig. 1 the pore pressures in Test Area I1 had decreased in 1961 to about the same value as that outside the loaded area indicating the completion of the primary consolidation. The scatter of the measured values outside the test areas is about &0.4 m for the three pore pressure gauges. In Test Area IV the piezometric level was still in 1969 higher than that of the surrounding soil. The difference was about 0.9 m. The consolidation of the clay below the gravel fill is still continuing (1970) in both test areas. The compression of the layers located between 2.5 and 7.5 m below the ground surface is since April 1959, about 16 cm in Area I1 and 23 cm in Area IV. However, the settlement rate has decreased more rapidly in Test Area I1 than in Area IV. Mainly secondary settlements have taken place after 1961 in Area I1 and the settlement rate of the clay layer mentioned above was in 1969 about 0.3 cm/year. At the same time the settlement rate for the same layer in Area IV was about 1.2 cm/year. Obviously the consolidation for this test fill is still in its primary state. Hansbo (1960) calculated in his report the settlement due to primary consolidation for the clay layer located between the depths -2.5 and -7.5 m. From the compression characteristics determined from laboratory consolidation tests he calculated a settlement of 54 to 57 cm for Test Area 11. In the calculations the effects of the disturbance caused by the driving of the drains have been neglected. Calculations by Hansbo based on the remaining measured excess pore water pressures in comparison with the observed settlements up to April 1959 indicated a primary consolidation settlement of 60 cm. From Fig. 1 it can be seen that the observed settlement in the middle of Test Area I1 was about 61 cm at the beginning of 1962 when the excess pore water pressures had dissipated. Thus the observed settlement is in good agreement with that calculated. In Test Area IV the settlement due to primary consolidation of the clay layer located between 2.5 to 7.5 m below the ground surface