Lateral Loading on Piles

For piling foundation, the lateral force on the pile head is resisted by the surrounding soil.

Simplified calculation can be carried out in accordance with recommendation by Tomlinson, M.J..
On the other hand, the pile and the surrounding soil could be modelled as frame elements with soil spring supports (ie. subgrade reaction method).

The soil spring could be determined in accordance with Figure 9 of Geoguide 1 (2nd Edition) published by Geotechnical Engineering Office (GEO). The spring constant (k_sj) is given by

k_sj = n_hz_jDz_j,

where
n_h = Constants of horizontal subgrade reaction
z_j = Depth of spring
Dz_j = Length of pile served by spring

Hence we can observed that the soil spring is depended on (i) n_h value of the soil; (ii) depth of soil spring; and (iii) distance between soil spring. It shall be noted that the value of soil spring is independent of the pile dimension. Typical value of nh value can refer to Table 6.11 of GEO Publication No. 1/2006- Foundation Design and Construction (2006) .

The reaction on the surrounding soil is assumed to be within the soil capacity, provided that maximum deflection of the pile/ surrounding soil is within 25mm.

The model can be carried out by linear analysis as the piles are surrounded by soil in all direction. If sheet piling/ pipe piling are analyzed with soil spring, non-linear analysis is required such that the soil element on the un-excavated side could only take up compression.

Soil Parameter

Soil Properties

The design for foundation and ELS works require some soil parameter like c’ (cohesion) and f’ (angle of shearing resistance) .

The soil parameters could be determined by triaxial compression test of soil sample obtained in ground investigation work.

Mohr’s Circle Diagram for different soil sample shall be as follow:

The triangles formed by the Mohr’s Circle shall be similar.

i.e q_i / (p_i + c’/tan f') = sin f’= constant
or q_i = p_i sin f’+ c’ cos f’

Hence, for same type of soil, a "p-q" graph can be plotted based on the laboratory test result, where p = (s₁ + s₃)/2 and q = (s₁ - s₃)/2.

We could then obtain the best fit curve for the data in form of q_i = mp_i + b. The soil parameters are then determined by f’= sin^-1 m and c’ = b/ cos f’