Geotechnical News • March 2012
45
expansion results in no pore pressure
change. My interpretation, using the
ideas presented in these articles, is as
follows:
Bearing in mind that it is the cone tip
which is moving and not the soil, the
fact that there is no change in pore
pressure in response to soil-structure
deformation can only be because
there is no change in void ratio during
deformation. This is consistent with
the classic case of constant volume
straining. My opinion as to how sand
could end up in this rather unique state
of packing is that these sands were
placed by bedload transport: a deposit
moved so far and so often that it now
exists at the constant volume void
ratio. This suggests to me that these
sands, although soft, are not liquefi-
able.
Slope stability
Where steady state seepage condi-
tions prevail within a natural slope,
or on the downstream side of an
embankment dam, we expect to see
a loss of hydraulic potential as we
go downslope. Since this reflects the
water energy lost to particle drag
forces as the water moves through
a stable soil-structure, this is as it
should be. This downhill drag is the
destabilizing influence trying to flatten
the slope. After allowing for seasonal
alterations in differential head across
the system, we don’t expect to see
that hydraulic flux change over time.
And that is what we hope to see from
any piezometers we have installed
for monitoring the slope. A pressure
distribution other than the established
pattern would indicate either a change
in the permeability of the section
(soil erosion), or be a warning sign of
movement within the slope.
The upper part of Figure 19 is an
instance of steady state seepage
through a stable soil-structure. It is
the lower part of this sketch, based
on the ideas introduced here, that
provides some additional insight into
what might be going on within the
slope. This applies more to natural
slopes which are often composed of
fine grained soil, and thereby prone to
a larger viscous component of energy,
than to the coarser soils used in
earthfill embankments. A piezometer
will not see the viscous drag forces
pulling down the slope, it will only
show the pressure component. But in
either slope, a change in piezometric
head, not attributable to changing
potential difference across the slope,
is a definite warning sign: And this
holds true whether the head increases
or decreases.
Ground improvements
Once we acknowledge the fact that
escalating pore pressures are a result,
and not the cause, of soil-structure
contraction or collapse, then it comes
time to look again at what we think
we are doing when we install verti-
cal drainage devices in the ground to
enhance the groundwater’s natural
drainage. Certainly, in the case of
non-granular compressible strata,
we hasten ground settlement by such
means as wick-drains. And that is
a good thing. We are venting the
pore pressures which are resisting
and retarding downward movement.
And when we install similar vertical
drainage elements in what is feared to
be liquefiable sands, it is exactly the
same thing we accomplish: We speed
up post-failure settlement. And that’s
about all.
Soil grains are not spheres
The numbers of particles of gravel, of
sand, and of silt required to make up
one cubic centimetre of soil are: One
single piece of gravel would do it;
40,000 sand grains would be needed;
and, for silt, the number is a stag-
Figure 19. Pressure gradients in seepage and relative motion.
