1

Simon Clarke on the
detective work needed before
 
repairs could begin
THE RESTORATION OF
EROS
In August 1990, vandals climbed onto
the statue of Eros in Piccadilly Circus,
swinging on the outstretched leg. The
statue suffered cracking in the calf of the
supporting leg.
Eros was commissioned in 1893 and
has required repair at various times in the
past, the most recent in 1986 when
Henshaws carried out a welding repair.
An initial examination carried out on
site revealed a large crack spanning over
50% of the circumference in the calf of
the supporting leg. Associated with the
crack were a large number of dowels and,
at the rear of the leg, a large aluminium
core plug, approximately 30 mm in
diameter; these had been part of earlier
repairs. A core plug was carefully removed
and examined in the laboratory, which
revealed that the statue consisted of an
outer aluminium skin, approximately 10
mm thick, and an aluminium fill material
that had been cast around the iron
armature supporting the statue. A series
of craze cracks was also spotted on the
front of the thigh of the supporting leg.
Liquid penetrant inspection revealed that
this cracking was just above the Roman
joint. From the nature of the cracking it
appeared to be in the form of a burst, that
is, pressure from behind rupturing the
surface and forcing it outwards. The core
hole was repaired temporarily using an
aluminium core plug.
The core plug sample revealed the
outer wall to be an aluminium alloy with
a low alloy content that had fractured
through the complete wall thickness in a
tensile mode with minimal elongation.
The infill material was an aluminium
silicon alloy, of around eutectic
composition, ie approximately 12%
silicon. This material was full of large laps
and oxide films and there did not appear
to be any bonding between the infill and
the outer skin. Microscopic examination
of the skin material showed it to be full of
oxide films and inclusions and possibly
some heat affected material, such as that
at the edge of a weld repair. This may
indicate that the failure had occurred
adjacent to the Henshaw repair.
English Heritage reported that both
the thigh and calf areas of the statue had
suffered similar damage in the past which
had been repaired by welding of the calf
and welding a plate in the thigh. It was
thought that although the main part of
the statue was hollow the armature
extended up to the shoulder to provide
some support over its full length. Infill
material in the supporting leg had been
added in the early 1930s, following
similar damage, in an attempt to
strengthen it. The statue had been cast in
the early 1890s, using a nominally pure
aluminium that was as pure as the
aluminium could be refined at that time.
The aluminium silicon infill material has
a much lower melting point and hence
would have less effect on the statue when
being poured in. However, because of the
contamination and debris in the statue
from the original core material, the fill
material did not fuse or bond to the skin
and hence the extra support gained was
only marginal.
It was recommended that the statue’s
calf should be repaired by welding and
the patch in the thigh replaced. Also
suggested was the possibility of filling the
statue so that stresses from the upper part
of the statue could be directly transmitted
to the armature, relieving the problem
area in the calf.
English Heritage wished any repairs or
modifications to the statue to be
reversible and to involve as little
disruption to the original work as
possible. With these conditions in mind
it was recommended that welding should
be carried out using compatible
consumables, ie pure aluminium, and
that any infill of the statue would need to
be easily removable for which fine grain
sand, compacted by vibrating, would be
appropriate.
Radiographic inspection revealed that
the armature had been incorporated into
the leg during casting but that it only
extended as far as the top of the Roman
joint in the thigh of the supporting leg.
Trials indicated that although the
armature was securely located at the top
of the thigh, it would obviously be of
little value for using as an infill material
to provide further support. Radiography
also showed that the crack appeared at the
change in section from the solid casting
in the foot and ankle area and the hollow
part of the leg. There were many foreign
particles present, including possibly
remnants from the original core materials
and large voids in the infill material.
Detailed ultrasonic examination


Penetrant testing of welded repair on supporting leg
(horizontal).
CONTEXT 42
23

1