Geotechnical News • September 2019
23
GEO-INTEREST
The Phoenix Ground Improvement Machine
Bill Hodge
My purpose in writing this article is to
bring to the attention of the geotech-
nical fraternity the fact that we now
have at our disposal a tool which can
turn loose saturated granular materi-
als
into highly competent
foundations.
Using this hardware, materials such as
sands, silts, and rock flour (slimes) can
be compacted into a highly dense and
dilative aggregation at depths of up to
about 20m below ground level. Field
records from three different sites are
presented below to substantiate this
claim. It is by combining into a single
poker both vibration and water drain-
age that this is accomplished.
During the 1980s the petrochemi-
cal industry was anxious to tap into
the Amauligak oil fields beneath the
Mackenzie Delta in the Canadian
offshore Arctic.
Gulf Canada Resources [
GCR
] was
the oil company who initially engaged
me to see if there was a sand island
solution to getting their rigs into
deeper water. Before finishing that
work
GCR
made a decision to switch
their focus to a hybrid steel vessel
called the Molikpaq and asked me
to attend to the geotechnical side of
things. It consisted of a 70m square
annular hull with an enlarged moon
pool into which sandfills would be
dredged; the idea being that the plat-
form would gain its lateral stability
by virtue of the frictional resistance
between the underlying deltaic sand
and that within the open-bottomed
moon pool.
The design criteria involved a horizon-
tally applied ice loading (70,000 ton,
pulsating monotonically at 2Hz), see
Fig 1. Because of this my main pre-
condition for the vessel’s deployment
was that the sandfill be densified to
prevent subsequent liquefaction of the
core. This was an unpopular require-
ment as it added a time consuming
complication. As it turned out, unbe-
knownst to me, the Molikpaq went to
work for a couple of years without its
core having been densified. Appar-
ently this was on the advice of a third
party who based his judgement on
centrifuge model testing. That is until
1986 when the core did in fact liquefy
under ice-loading while the platform
was drilling on location. Then
GCR
asked Phoenix Engineering Ltd [
PEL
]
if we could fix it.
Restabilization of Molikpaq
Blast densification was an obvi-
ous option. But detonating multiple
charges of TNT inside the core of an
offshore platform while it was drilling
into an oil field cause some safety
and environmental concerns with the
regulating authority, Canadian Oil and
Gas Lands Administration [
COGLA
].
So while licensing discussions went
on
PEL
looked for a mechanical
solution. Since there was only 2.5m
(11ft) of height between the surface of
the sandfill and the underside of the
platform’s steel deck (see Fig 2), the
standard deep compaction pokers such
as Vibroflots were ruled out.
It was then that the idea of the vibro-
drain, hereinafter referred to as the
Phoenix Machine [
PM
], occurred to
me. By simply combining the ben-
efits of the seepage forces (Ref 1) we
knew about from our earlier research
at the hydraulic laboratories of NRC
in Ottawa had examined at NRC with
some sort of vibrator seemed to be
worth trying; vibration being the best
approach to improving the density of
granular (non-cohesive) loose materi-
als. The mechanical device we came
up with to produce vibrations was an
eccentric weight rotated by a custom
Figure 1: Molikpaq under ice attack.
