Basic HTML Version
t h e b u i l d i n g c o n s e r v a t i o n d i r e c t o r y 2 0 1 2
1 5
1
Professional services
inherent in this approach, not least:
• introducing new functions and spaces
without damaging the integrity of the old
• controlling damp ingress to an acceptable
level without changing the aesthetic of
the castle
• managing safe access and egress without
introducing fire escapes
• accommodating new hidden services
runs within and without
• minimising future maintenance,
particularly at high level
• defining the junctions between new
and old.
Conservation and
construction issues
The liberal use of concrete renders from the
early 20th century restoration works has
not aided the survival of the castle, which is
recorded as having long-standing problems
with water ingress and damp. These were
serious enough to cause some occupiers to
leave. The application of an asphalt coating to
the flat roofs and parapet walks helped but,
as with the use of cement-based pointing, it
proved to be a short-term fix at best.
Encased in cement by Viscount St
Davids and in glass fibre by later owners,
permission was given at an early stage to
allow the removal of linings, cement renders
and pointing. This would allow a better
understanding of the story of the development
of the castle and its condition, prior to an
application for listed building consent.
As work progressed, the extent of
the damage caused to steel beams and
unprotected reinforced concrete (RC) work
became apparent. Roofs were formed from
concrete cast in situ onto vaulted corrugated
steel sheets between RSJs. The breakdown of
the asphalt covering had severely damaged
the corrugated steel and the top flange
of the RSJs. The RC floor slabs had been
built directly into the damp walls and
had been cast with too little ‘cover’. (Over
the years, dissolved carbon dioxide from
the atmosphere progressively reduces the
alkalinity of the concrete which protects the
steel reinforcement. The thickness of the
concrete over the steel is therefore critical
to its durability.) Concrete analysis showed
that in all of the concrete floors, the depth
of carbonation exceeded the concrete cover
to the reinforcement, suggesting that all of
the reinforcement was by this time liable to
ongoing corrosion.
The engineer’s report concluded that the
extent of the damage to the reinforcement
and the structural steel, and the degree to
which it had been built into the masonry,
meant that in situ repairs would be
impractical.
On the other hand, the replacement of
the floors and roofs would mean the loss
of all the 1950s alterations and much of the
Viscount St Davids restoration of 1910. There
was a general consensus that the 1950s work
was of little architectural or historical value.
The archaeologist had already concluded
that the 1910 alterations had not adhered to
any of the original floor levels or staircase
locations and most of Viscount St Davids’
architectural details had been lost during
the later work. The significance of his
involvement, certainly as far as the Society
for the Protection of Ancient Buildings
was concerned, revolved around what was
perceived as an early use of in situ concrete
work in west Wales.
Justification for removal rather than
repair of the RC structure by the architect
was based on:
• the identification of much better
examples of early RC work in west
Wales such as those at Caldey Island
Monastery (1912) and the Towy
Works, Carmarthen (1907–9)
• the long-term maintenance implications
of not replacing the floors
• the likely effects of any collapse and
the resulting damage to the medieval
fabric should corrosion continue.
After a prolonged debate, consent was
eventually granted. It was conditional on
the specific testing of each slab and the
development of a method statement for
the removal of the floors (from the bottom
up) aimed at preserving the integrity of the
castle walls.
Intervention
Repairs and alterations were designed to
ensure that the surviving medieval fabric
was protected and consolidated, and new
work was designed with reversibility in
mind. Structural components such as roofs,
for example, were attached to the structure,
not built into it, so that if necessary the
new components could be removed with
minimum disruption to earlier fabric.
The external repair of the medieval fabric
using matching Pennant sandstones and
hydraulic lime mortars has been extensive and
time consuming. Internally, the masonry walls
have been plastered with hemp lime plaster.
Last repointed in sand and cement in the
early 20th century, all the pointing has now
been renewed in hydraulic lime (NHL 3.5) and
sand mortar. The main body of the medieval
tower has been hacked out to an average depth
of 60mm and pointed in three ‘passes’. The
final coat was trowled and brush finished
over much of the rubble face work in what is
locally known as a ‘parged’ finish. In addition
the 15th-century square tower on the exposed
south face has received two shelter coats of the
same mix, brush applied. The roughly coursed
masonry of the 20th-century annexe has been
flush pointed in two passes.
To avoid confusing the history
of the building, the internal fit-out is
uncompromisingly modern, incorporating
polished limestone floors, glass screens
with aluminium frames and bespoke
joinery. In the one location where the new
interior extends outside, in the form of a
structure which tops the annexe roof (see
title illustration), the modern vocabulary
is continued, although it interacts with the
original fabric by reflecting the D-shaped
form of the original tower.
This dialogue between new and old is
continued internally through the use of hemp
lime plasters for many of the new suspended
ceilings and partitions.
Completed in the autumn of 2011, it is
hoped that the quality of the repairs and
alterations will sustain the next chapter
in the life of this important building, both
physically and economically.
Peter Holden
RIBA AABC is managing
director of Acanthus Holden Architects (see
page 19) and was director in charge of the
project with direct responsibility for the
conservation of the medieval castle.
James Meek
MIfA is head of field services
at Dyfed Archaeological Trust and was
responsible for the archaeological research
for the project.
View of the castle from the south-east
New lime render on the staircase walls