Geotechnical News December 2010
57
GEO-INTEREST
at rest and ended up at a lower eleva-
tion, the particles must at some stage
have being accelerating downwards.
That means they were experiencing
some modification, let’s call it “a”,
superimposed on the original gravita-
tional field “g”. They were in fact un-
der the influence of a “g minus a” flux.
Therefore, the weight of the beakers as
measured by the scales should be:
Before
m g
During
m (g – a)
After
m g
And this quite definitely would
prove the stable conditions were equal
in weight and the “during” beaker
lighter than both. Fortunately, how-
ever, things are not quite as simple as
that: In what I’ve done so far I’ve been
ignoring the water !
This introduces a few complica-
tions: The fact is that the solids mov-
ing down had to be accommodated by
an equal volume of water moving up,
and at some stage the water had to be
accelerating too. Water density is only
about 60% of buoyant grains and this
argues in favour of the active beaker
still being lighter. I simply don’t know
the ratio of the accelerations. To add to
the murkiness of the situation, don’t we
all know failure is accompanied by an
increase in pore water pressure when
a contractive structure collapses ? So
could water pressure on the base of the
beaker make up the difference caused
by the descending solids, and just add
up to making everything turn out the
same in the end ?
As we engineers know, thinking
about problems only gets you so far,
eventually you need to step into the
real world of a site to get the answers
to what really might be going on. Obvi-
ously it is now time for a reality check
by a real life enactment of the “thought
experiment”. The problem here is that
a laboratory test would be quite a dif-
ficult experiment to perform since it
would involve some way of introduc-
ing a jolt enough to cause failure with-
out the attendant commotion upsetting
the vertical reading. Next best thing
would be to reduce the test to its bare
essentials and see if I could find a way
of doing the measurement with what I
could find around the house.
Kitchen Experiment
Looking at the “before” and “after”
beakers it is apparent that mainly
what changed was the position of the
centre of gravity of the particles; it is
lower after the collapse than before. So
perhaps a very simple test involving
just one solid particle would tell me
something about what might be the
range of possibilities in the “during”
beaker. And this setup was so simple
that I found in my kitchen enough for
a “quick and dirty” version of such a
test. Figure 2 shows all that’s required.
Setup to cleanup takes about half an
hour.
To have enough time to see what
was happening I needed to arrange to
keep the action slow, and at the same
time to produce a weight (buoyant)
heavy enough to show up on my scales.
After trying a few things which worked
well enough, like a small potato and an
egg (hard-boiled for obvious reasons),
I settled on a golf ball.
The procedure was to hold the ball
by a wire thread just below the water
surface in a tall clear plastic container
(spaghetti jar). Then, after taring/zero-
ing the readout, let the ball fall while at
the same time watching the reading on
the scales. Right away I had what I was
looking for. What I saw was that the
scales showed nothing much until the
golf ball had fallen about 10 cm – then
it showed the full buoyant weight of
the ball (~ 5 grams), and this appeared
at a stage where and when the ball was
still far above the bottom. This dem-
onstrated quite clearly that the weight
of the particle was felt when it had no
hard physical contact with the scales.
Two clear and undeniable conclusions
are:
1. that the weight of the ball was trans-
mitted to the scales by a column of
pressurized water under the falling
ball, and
2. that pressure transmission required
some amount of movement by the
ball through the water.
Figure 2. The kitchen experiment.
