4.3 TEST RESULTS ON END-BEARING SINGLE PILE
4.3.6 Stage 6: NSF due to Surcharge
Proceeding with the test procedure, in-flight sand surcharge was subsequently applied onto the soil surface without stopping the centrifuge. This was achieved by pushing out the base plate beneath the two sand hoppers to align vertically the holes in the base plate with those at the base of the sand hoppers to discharge the sand into the model container, as explained in detail in Chapter 3. The flow of sand was further diverted by the guiding plates installed beneath the sand hoppers with adjusted angles for overcoming the Coriolis
113
effects and distributing the sand evenly in the container. The 9 kg of sand discharged onto the soil surface was equivalent to about 45200 kN in prototype scale. The surcharge intensity was thus about 40 kPa, considering that the internal size of the soil container was 420 mm by 420 mm in model scale, or 33.6 m by 33.6 m in prototype scale.
Fig. 4.19(a) shows the development of excess pore pressure after the application of surcharge, taking reference of the PPT readings immediately before the application of surcharge. It can be seen that the 40 kPa surcharge initially caused an excess pore pressure increment of about 40 kPa, as expected in an undrained condition. The excess pore pressure then dissipated with time and essentially fully dissipated after 830 days. Fig.
4.19(b) zooms in on the development of the PPT readings for the initial 5 days upon the application of the surcharge, which clearly shows that the surcharge loading of 40 kPa was essentially applied expeditiously within 0.5 day. Fig. 4.20(a) shows the development of incremental soil settlement after the application of surcharge. The soil settlement within the first 5 days upon the application of surcharge is shown in Fig. 4.20(b). It can be seen that the expeditious build-up of 40 kPa surcharge caused an immediate soil settlement of about 35 mm within half a day, compared to only 1 mm immediate settlement of the pile as shown in the figure.
Recalling that the application of permanent dead load in the previous stage has caused the NSF along the pile shaft to reduce by a certain degree to below the values when fully mobilized, it is speculated that the expedient large downward movement of the soil relative to the pile within half day upon surcharge should have fully mobilized the available NSF between the soil and the pile in an undrained manner with constant effective stresses. This postulation can be readily verified by examining the load
114
increment along the pile shaft upon surcharge. Taking reference of the loads immediately before the surcharge, the development of incremental loads with time after surcharge is shown in Fig. 4.21(a), with the immediate responses within the first 5 days zoomed in and shown in Fig. 4.21(b). Consistent with the quick build-up of surcharge and large immediate soil settlement within half day upon surcharge, significant increment of forces was observed to develop on the pile within half day upon surcharge as shown in Fig.
4.21(b). The load transfer curves as plotted in Fig. 4.18 clearly shows that the loads along the pile shaft increased substantially along the whole length of pile shaft within half day upon surcharge. By assuming that all this load increment occurred in an undrained condition and the effective stresses in the soil remained the same as that prior to the application of surcharge, the β-curve thus derived fits the test data reasonably well using a consistent β value of 0.24. It is noted that the load on the pile head at the original ground elevation also increased slightly above the 700 kN dead load applied at Stage 5, which should be attributed to the downdrag load of the surcharged sand shrouding the pile shaft above the original model ground surface.
The observation of very large mobilization of NSF along pile shaft upon surcharge under undrained condition is in line with that observed by Little (1994) in a field test in which the NSF on a free-headed pile group was induced by a surcharge of 4-m embankment, as reported in Chapter 2. The 4-m embankment was constructed within 4 days during which period the clay was deemed to be in undrained condition due to its low permeability. However, upon completion of the embankment, very large of NSF was observed to mobilize along the pile shaft which amounts to about 50% of the final dragloads at the end of the consolidation about 2 years after the surcharge.
115
The subsequent consolidation of the clay after surcharge will increase the effective stress as well as shear strength of the clay. As shown in Fig. 4.18, the derived β value, using the increased effective stress at the end of the consolidation, for fitting the long-term load distribution along the pile shaft was 0.24, which is consistent with those derived in the previous stages. If α method was used to back-analyze the test data at the end of the consolidation, the derived α value will differ greatly depending on whether the increased shear strength or the in-situ shear strength is used (See Fig. 4.2 for the two measured shear strengths). The derived α value based on the increased shear strength after soil consolidation was about 1, which was consistent to those values derived in the previous stages as well as with data reported by other researchers like Walker and Darvall (1970), Endo, et al. (1969) and Leung et al. (2004). For the case that in-situ shear strength was used, as likely in practice, the derived α value would be as high as 1.9 in order to match the test data as shown in Fig. 4.18. Evidently, the common practice of using in-situ shear strength measured before the commencement of a project at the design stage would grossly under-estimate the dragload on piles when the soil is subjected to substantial surcharge and consolidation which causes substantial increase in shear strength of the clay.
In this case, using effective stress method with typical β value and updated effective stress after surcharge provides a more reliable way of evaluating NSF on piles than total stress method which requires the measurement of undrained shear strength in the long term after the consolidation finishes.
It should be noted that the much larger soil settlement under the surcharge loading caused the NSF to be almost fully mobilized even near the neutral point. As such, the calculated maximum dragload at the NP is much closer to the measured data, as can be
116
seen from Fig. 4.18. This is in contrast to the previous stages when the soil settlement was substantially less and the mobilization near the NP is far from fully mobilized, leading to a much larger discrepancy between the calculated maximum dragload at NP and the measured data. The intensity of surcharge loading as well as other factors such as pile slenderness ratio and pile-soil stiffness ratio affect the degree of mobilization around the neutral point and in-depth analysis on this aspect is highly desirable. This will be explored with the aid of numerical analysis presented in Chapter 6.