Page 23 - Numerical Analysis and Modelling in Geomechanics

4 JOHN W.BULL AND C.H.WOODFORD
into the void. The radius of the column is approximately equal to that of the void
[24]. If the medium is a subgrade, the detonation gases dry the void and in time
the walls of the void slide down, disturbing the subgrade. The shape of the
disturbed subgrade is conical with its base upward [14, 25, 26]. The cone apex
angle at the point of detonation is between 82.14° and 93.74° [14, 25].
In sands, following a detonation there is either immediate surface ground
heave or small surface settlement followed by continued ground settlement for
about an hour. Consolidation of the sand occurs over several hours with water
escaping to the surface. The results of standard penetration tests or cone
penetration tests taken immediately following detonation will be misleading as
the test results will indicate little effect of the detonation, although the clearly
visible large surface settlements make clear that considerable density increases
have taken place [27].
Clay subgrades have high compressibility, high plasticity and low filtration
properties that ensure that voids caused by an underground detonation are almost
hermetically sealed. Full-scale experiments have shown that in clay large cavities
are distinguished by low stability, but that if these voids are spherical with a
diameter of up to 6 to 7.25 metres, they can be relatively stable [28]. To produce
a camouflet with λ =−1.388 having a void diameter 7.25 m would require a 333
c
kg mass of TNT (1.048W 0.333 ). The probability of voids being formed near the
surface in granular materials is negligible, but the probability does increase as
the depth of detonation increases [16, 17].
Test data for camouflets where there is little or no surface disturbance is
scarce [25]. When a camouflet is formed, at the air-ground interface there may
be no disturbance, a small mound, a depression, a hole or only loose subgrade.
The shape of the disturbed subgrade above the camouflet is, as previously
described conical, base upward [25]. No specific research has been carried out to
determine the apex angle of the cone, although twice the subgrade’s angle of
repose has been considered as representative [29]. There is no published research
on the diameter of the surface disturbance related to the camouflet size and to the
depth of detonation.
At the formation of a camouflet, the detonation produces a shock wave. On the
shock wave front the subgrade is compressed, while behind it the subgrade
expands. When the compressive shock encounters the air—ground interface, a
negative stress wave is generated that propagates back into the subgrade. At
certain depths, the sum of the two stress waves equals the dynamic tensile
strength of the subgrade and pieces of subgrade break away. This produces a new
free surface and even more pieces of subgrade break away. At increasing
distances between the point of detonation and the air-ground interface, the
pressure decreases until it does not exceed the tensile strength of the subgrade.
Ultimately, the surface disturbance is only a small elastic movement, with little
disruption of the subgrade layers and with surface subsidence occurring perhaps
later [30, 31].