CEN/TC 250/SC 9 N 560

CEN/TC 250/SC 9
Eurocode 9: Design of aluminium structures

Email of secretary: [email protected]

Secretariat: SN (Norway)

N560 GARIGLIANO PROHITECH

Document type: Other committee document

Date of document: 2017-09-24

Expected action: INFO

Background:

Committee URL: http://cen.iso.org/livelink/livelink/open/centc250sc9

The use of aluminium in the restoration of the “Real Ferdinando” bridge
on the Garigliano river
F.M. Mazzolani
University of Naples “Federico II”, Naples, Italy

ABSTRACT: Starting from the experience of the retrofitting of old suspension bridges in France by means of
aluminium structures, a similar solution has been proposed and used in the structural restoration of the “Real
Ferdinando” bridge on the Garigliano river near Naples, which was the first suspension bridge built in Italy in
1832. The historical background, the structural restoration criteria, as well as the design and the execution of
this bridge are described in this paper.

1 INTRODUCTION led to the possibility of increasing the weight of the

road vehicles, while preserving both the existing ca-
The use of aluminium alloys is today an emerging
bles and piers without significant strengthening
activity which compete with steel in many specific
(Figure 1).
fields of structural engineering (Mazzolani, 1995). A
new field of particular interest is the one of structural
restoration, where aluminium and its alloys represent
a new material whose properties can be exploited in
very promising way. An interesting experience has
been done in the restoration of old suspension
bridges in France during the seventies (Mazzolani &
Mele, 1997). These structures, built during the 19th
a) b)
century, were made of wooden deck, masonry piers,
Figure 1.The Montmerle bridge on the Sôane river in France:
steel girders and steel suspension cables. Their bad a. Two spans of 80 meters; b. The aluminium longitudinal gird-
conditions required structural consolidation er.
operations, also in view of re–using these bridges in
a new improved road planning. After a national In the retrofit of the Groslée bridge (a single 174 m
competition, in which both steel and aluminium bay) the new deck structure is made of three longitu-
solutions were presented, the French Road dinal aluminium truss girders, connected above to a
Authorities accepted the adoption of an aluminium light reinforced concrete slab (Figure 2).
deck in the retrofitting project of three bridges (the
Montmerle and the Trevoux bridges on the Sôane
river; the Groslée bridge on the Rône river). They
were completed within 1975.
The ECCS Committee on the “Design of Aluminium
Structures” visited these bridges in 1978 and his
members were very favorably impressed by these
structural solutions, which thanks to the prerequi- a) b)
sites of aluminium alloy (lightness, corrosion re- Figure 2. The Groslée bridge on the Rône river in France:
sistance, rational shapes of structural members due a. One span of 174 meters; b. The piers and the aluminium
to extrusion) allowed to preserve as much as possi- deck.
ble the old structural elements (like masonry piers
and cables), giving rise to very effective results in This structural scheme can be remembered as a re-
term of both cost and structural performance. In the markable example of co-operation among different
Montmerle bridge (two 80 m bays) the use of alu- structural materials, each of them utilised in opti-
minium both for the two longitudinal truss beams mum working conditions: the old masonry piers, the
with bolted connections and for the deck structures high strength steel suspension cables, the stainless
steel suspension ties, the aluminium alloy truss gird-
ers with high strength steel bolted joints, the light re-
inforced concrete slab, the casted aluminium ele-
ments for connecting suspension cables and ties.
At the end of eighties, a structural restoration project
of the oldest Italian suspension bridge, the “Real
Ferdinando” bridge on the Garigliano river (1832), a) b)
has been proposed in the context of a wider rehabili- Figure 3. a. A painting of the “Real Ferdinando” bridge on the
tation program of that area, which wanted to re-use it Garigliano river (Filangieri’s Museum, Naples); b. a paint-
ing of the “Maria Cristina” bridge on the Calore river
for a pedestrian connection (Mazzolani, 1990). (Filangieri’ s Museum, Naples).
Considering the positive french experience, it
seemed immediately that an aluminium solution The first Giura’ s idea was oriented to a chain sus-
would be very appropriate for the planned purpose. pension bridge solution, which was favourably ac-
But it was necessary to develop a comprehensive cepted by the king, who decided to send him to Eng-
comparison between the more classical steel solution land in order to visit the already existing bridges of
and the aluminium one in order to convince the pub- this type. During a three years journey, he carefully
lic Authorities (Landolfo et al., 1989). A long proce- observed the Union Bridge on the Tweed river, with
dure of about ten years followed the first approach to a span of 112 m built in 1820 according to the pro-
the problem and finally in 1998 the new aluminium ject of the captain of the Royal Navy Samuel Brown,
bridge has been erected (Mazzolani, 1998). and also the Bangor Bridge on the Menai strait in
This exciting history is told in the following, starting Wales, with a span of 177 m, designed by William
from the chronicle of the original design and con- Telford and under erection at that time.
struction made at the beginning of 19th Century, Coming back in 1828, Giura very quickly designed
through the motivation of the innovative choice in- his bridge on the basis of the recent english experi-
volving the aluminium use in the strict respect of the ence and referring to the Navier’ s studies on sus-
restoration criteria, until the acceptance of the new pension bridges.
design and the recent erection of the “Real Ferdinan- On 10 May 1832 the King Ferdinando II invited all
do” bridge on the Garigliano river. his Court early in the morning on the left side of the
Garigliano river at the border with the Papal States
2 HISTORICAL BACKGROUND for the opening ceremony of the new “Real Ferdi-
The end of 18th Century and the beginning of 19th nando” bridge. This bridge was an old big aiming of
Century was a challenging period in the Bourbon the King, which was satisfied by the Giura’ s work.
Kingdom of the “Two Sicilies”, where important ac- In fact, as it is written in the Annual of the Kingdom
tivities in the field of new technologies were devel- of the Two Sicilies, this bridge was the first bridge
oped. One can remember, for instance, that the first of this type built in Italy near the Formia’ s fields,
railway in Italy was the Napoli – Portici connection. where Vitruvio Pollione – the “ Prince of the Archi-
The Bourbon iron mills, so-called Ferdinandea and tecture ” - was born.
Cardinale located in Calabria, beside to guns and Some data on the construction of the bridge were
weapons, produced for the first time casted iron ele- found on the old papers. The design, manufacture
ments for civil engineering applications, like pre- and erection phases took four years all together. The
casted columns of different shapes (Mazzolani, cost was 75.000 ducats, less than one half of the ex-
2001). pectation! The structural weight was 70.000 kg of
The structural elements for the first suspension iron coming from the Calabria mills. The piers were
bridges, like chains and anchorage, were produced made of hard calcar stone able to carry a load of 600
there in relation with the construction of the two kg/cm2.
bridges on the Garigliano river and on the Calore The chronicles of such time tell that the King in the
river in Campania (Figure 3). middle of the deck ordered that two companies of
In the period between 1780 and 1824, several struc- lancers pass at a steady trot on the bridge, together
tural solutions were proposed for the construction of with sixteen trucks of artillery. The performance of
the bridge on the Garigliano river, made of masonry the bridge was very good according the expectation:
with one or two arches, or made of iron and timber. Naples and Rome were more close.
But no one proposal was approved, because of both Three years after, a second suspension bridge the
the low contents of the technical solution and the Maria Cristina Bridge on the Calore river, was erect-
high cost. Only in 1825 the King Ferdinando II de- ed, again designed by the same engineer Luigi Giura.
cided to charge the neapolitan engineer Luigi Giura, The “Real Ferdinando” bridge was in good shape for
expert in roads and bridges, to design the Garigliano more than one Century until the world war two (Fig-
bridge. ure 4).
 Historical Record Office of Naples, representing an
extraordinary and interesting set of information
(Mazzolani, 2006).
The structural scheme of the suspension bridge is
characterised by a deck 5.80 m wide with a span of
85 m (Figure 6).

Figure 4. The “Real Ferdinando” bridge between the two world

wars (photo courtesy G. Matacena).
Figure 6. The graphical reconstruction of the original “Real
Ferdinando” bridge.
In fact, in October 1943 the deck of the bridge was
mined and destroyed by the pioneers of the german
army in order to slow down the advance of the allies. The suspension system consists of two couples of
But it was quite unuseful, because they built very chains, each chain being composed by four elements
closely an arch bridge which is still in function. of rectangular cross-section made of puddled iron,
Very recently a new cable stayed bridge has been which are connected together by means of cylindri-
erected not so far in that zone, where the traffic be- cal pins (Figure 7 a). The vertical suspension ties,
tween the Campania and Lazio regions is very 1.36 m spaced, are connected to the pins of the
heavy. chain. They inferiorly hang the longitudinal iron
Until the end of 80’s only the four piers of the “Real girders with rectangular cross-section, which support
Ferdinando” bridge stayed in place (Figure 5), to- the wooden deck made of transversal oak beams,
gether with the access roads on the banks and the 1.36 m spaced as the vertical ties, completed by a
lateral iron chains with the four characteristic wooden deck planking. The four masonry piers are
sphinxes above the anchorage points of the suspen- conceived in “Egyptian style” with floreal top capi-
sion chains. All these remaining parts were in bad tals (Figure 7 b).
condition due to the injury of man and the ravages of
time.

a) b)
Figure 7. a. Detail of the pins connecting the iron elements of
a) the two suspension chains. b. The capital of the pier in “Egyp-
tian style”.

The supporting devices of the suspension chains are

located inside the capitals. These special devices
work like a pendulum, allowing the chain to have
b) c) horizontal displacements in the plan of the bridge
Figure 5. a. The four piers of the “Real Ferdinando” bridge af-
and transferring to the pier vertical actions only
ter the demolition of the deck; b. the remaining lateral iron (Figure 8 a). This was a very clever and forerunner
chains of the “Real Ferdinando” bridge before restoration; c. idea of the designer.
the position of the chain anchorage. The retaining chains are anchored in a massive block
of stone masonry at a depth of 6 m and at a distance
3 THE ORIGINAL DESIGN of 24 m from each pier, where the sphinxes indicate
the point of anchorage (Figure 8 b).
At the beginning of the restoration process in 1984, The analysis of the existing original material gave
the original design report and some drawings of the the possibility to identify many important data of the
“Real Ferdinando” bridge have been found at the design process. The total dead load of the bridge
structure, including chains, ties and deck, can be
evaluated in about 260 kg/m2 and the structure has In addition no consideration to the deformability
been checked for a live load of 240 Kg/m2. problems, which were neglected because of lack of
sensitivity. In fact, on one hand we have to remem-
ber that the suspension bridge is the descendant of
the liane bridge of the primitive people, and on the
other hand the deformability checks belong to the
modern philosophy of structural design, which only
in the last decades introduced the so-called “service-
ability limit state”.

4 THE STRUCTURAL RESTORATION

CRITERIA
a) b)
The structural restoration of the “Real Ferdinando”
Figure 8 a. The “pendulum” supporting the suspension chains bridge has been based on the choice to rebuild the
on the top of the pier. b. The sphinx indicating the joint of an- missing deck and to substitute the suspension sys-
chorage of a chain. tem, in order to obtain a pedestrian bridge connect-
ing the two parts of the new archaeological area cre-
The analysis of the existing original material gave ated around the Garigliano river.
the possibility to identify many important data of the The design of the new bridge structure has been
design process. The total dead load of the bridge done following three basic criteria:
structure, including chains, ties and deck, can be a. to keep the same image at the original bridge in
evaluated in about 260 kg/m2 and the structure has rebuilding the new deck with the suspension system;
been checked for a live load of 240 Kg/m2. b. to provide the deck with a sufficient stiffness,
For the total load of 500 kg/ m2 distributed along all both in vertical and horizontal directions, which was
the span, the value of the axial force in the chains is completely absent in the old bridge;
500 tons, which corresponds to a stress of about 15 c. to identify a solution based on contemporary tech-
Kg/mm2 in the material. This value seems to be too nologies and modern materials.
high for the puddled iron, also considering that the On the basis of the first criterion, it has been decid-
assumed amount of live load is very low in compari- ed:
son with the actual loading conditions. • to preserve the existing masonry piers by means of
From the complete structural analysis, the followings appropriate consolidation operations;
aspects can be identified: • to keep the same number of chains (two for each
• the structural scheme has a stable performance un- side);
der vertical loads, provided that they are symmetri- • to keep the same distance among the suspension
cally distributed; ties;
• the deformability of the bridge is very high and • to maintain the same width of the deck between the
produces a mechanism effect, when the live load is railings;
not symmetrically distributed (this result has been • to use a structural system for the deck as close as
confirmed by the chronicles of those time telling that possible similar to the original one.
a local peasant observed a raising of half meter at the On the basis of the second criterion, the longitudinal
edge of the deck when a carriage entered in the girders had to be sized in order to guarantee ade-
bridge from the opposite edge); quate vertical flexural stiffness, and integrated by
• due to the total absence of horizontal bracings, the horizontal cross-bracings in order to provide lateral
bridge is not able to resist any kind of lateral forces, stiffness in line with the modern codification.
like wind and earthquake, under which too large hor- According to the third criterion, the technologies of
izontal displacements can arise; the puddled iron and of the wooden deck have to be
• the considered live loads is inferior even to the substituted by the equivalent ones, which today can
ones which today are requested for a simple foot be identified in the wide family of the modern metal-
bridge. lic materials. In particular, the use of the following
These remarks confirm the general consideration materials has been examined:
that the old structures of the 19th Century were main- • the high strength steels under form of spiroidal
ly designed to resist vertical and symmetrical loading wires for cables, substituting the iron chains;
conditions, but unsymmetrical and horizontal load- • the low carbon steel for the deck structure and the
ing conditions were ignored, also because of the ab- suspension ties;
sence of adequate calculation methods, which have • the aluminium alloys, as on alternative to steels for
been set-up only during the second half of the 20th the deck structure, taking advantage of their perqui-
Century. sites of lightness and corrosion resistance.
As far as the new technology is concerned, some
possibilities have been taken into consideration:
• pre-fabrication of large elements built-up by weld- 5 THE NEW BRIDGE
ing in shop; After the choice of the aluminium solution and the
• field-bolting of pre-fabricated elements with full corresponding design, it was necessary to wait about
strength joints;
ten years for the construction. In 1998 the erection of
• vertical pre-stressing for consolidation of masonry the bridge has been completed. The deck of the
piers; bridge is made of aluminium alloy (type 7020 T5 for
• new bearing and supporting devices for cables. longitudinal girders and 6060 T6 for transversal
In this scenario of materials and technologies, two beams).
alternative solutions have been considered for the The two longitudinal girders have a structural
new bridge deck: one using steel and the second us- scheme of Vierendeel beam, made of two horizontal
ing aluminium, deciding to compare them in quanti- chords connected by vertical members having the
tative way for the final decision. same spacing of the suspension ties (Figure 9). The
A wide parametrical analysis has been performed in vertical members of each girder are split into two el-
order to identify the main parameters influencing the ements in order to receive inside the transversal
behaviour of a suspension structure by considering beams (Figure 10 a), whose extremities are connect-
both cases of steel and aluminium deck, but keeping ed to the suspension ties (Figure 10 b).
always the steel cables. The main result of this nu-
merical analysis shows that the repartition of action
between steel cable and aluminium deck in the struc-
tural model of a suspension bridge produces a stress
reduction in the steel cable greater than the simple
amount due to the deck lightness.
Following this analysis, for the given structural
scheme under the design conditions, the structural
analysis, the sizing, the strength and stability checks
have been performed in both cases.
The results put in evidence the main differences be-
tween the two solutions:
a. the dead load of the deck is 95 tons in the steel
solution and it is reduced to approximately one third
in the aluminium solution;
Figure 9 The new “Real Ferdinando” bridge after erection.
b. the suspension high strength steel cables have a
diameter of 60 mm in the steel solution, which is
reduced to 40 mm in the aluminium solution with a
total weight less than one half;
c. the aluminium solution does not require any kind
of protective coating, which contrary is es-
sential in the steel solution due to the humid envi-
ronment;
d. the greater lightness of the aluminium solution
reduces the actions on the piers, which therefore re-
quire less important consolidation operations;
e. transport and erection operations become easier
because the aluminium elements are lighter that the a) b)
steel ones; Figure 10 a. Detail of the mesh of the Vierendeel type girder.
f. last but not least, the two solutions practically b. Connection of the suspension ties to the transversal beam.
lead to the same cost, because the higher price of al-
The horizontal bracing system is made of X-bracing,
uminium is compensated by saving quantities in oth-
which are located in the mid layer of the bottom
er components, like cables, and in constructional op-
chord of the Vierendeel beam (Figure 11 a), accord-
erations, like pier consolidations, easy erection and
ing to a mesh equivalent to two-three spacings of the
coating.
transversal beams.
This last result has been determinant in convincing
The deck is covered in the central part by a wooden
the Italian Road Administration in the choice of the
floor which is longitudinally limited by a seating
aluminium solution, which has been considered in-
bench separating the lateral fiddley opening (Figure
novative and in line with the recent International
11 b).
Restoration Charts, which state that the new works,
beside the old ones, must have an unmistakable
The masonry piers have been consolidated by substi-
modern and reversible feature.
tuting the damaged stones with new ones (Figure 12
a). Also the four sphinxes has been cleaned and re- minium deck guarantees to the bridge a great relia-
stored (Figure 12 b). bility and durability.
The anchorage system has been revised and the sus-
pension supports on the top of the piers have been
substituted with modern devices. The old pendulum 6 FINAL REMARKS
system designed by Luigi Giura has been removed The challenging history of the “Real Ferdinando”
and located beside the bridge in a kind of museum, bridge stimulates some final remark.
where also the remaining parts of the old chains are Looking to the past, one can remember that the King
displayed to the visitors (Figure 12 c). Ferdinando II charged Luigi Giura to design an in-
novative bridge and sent him to England to study
new innovative technologies in order to fulfil the re-
quested requirements.
After more than 150 years, the innovative experience
of the restoration of the french bridges by means of
aluminium inspired and supported the technical and
a) b) cultural bases necessary to propose also in Italy this
Figure 11 a. Lateral support of the girders and horizontal cross- kind of solution.
bracings. b. The deck of the new bridge.
This background confirms that the history of the
“Real Ferdinando” bridge goes on by summing up an
original record to a new record. In 1832 the “Real
Ferdinando” was the first suspension bridge in Italy
and now at the beginning of the third millennium it
represents the first italian bridge made of aluminium.

REFERENCES
Landolfo, R., Mazzolani, F. M. & Mele, E. 1989. Riesame dei
a) b) c) ponti sospesi di Montemerle e di Groslèe: confronto fra acciaio
Figure 12 a. The masonry piers after consolidation and restora- e alluminio. Proc. of the XII C.T.A. Conference, Capri, Octo-
tion; b. The new suspension steel chains below the sphinx; c. ber.
The original “pendulum” device designed by Luigi Giura. Mazzolani, F. M. 1990. Il restauro strutturale del ponte “Real
Ferdinando” sul Garigliano. Costruzioni Metalliche 2.
The construction of the bridge has been done by re- Mazzolani, F. M. 1995. Aluminium Alloy Structures, E&FN
specting in general the design data. Only one sub- SPON, Chapman & Hall, 2nd edition.
Mazzolani, F. M. 1998. L’alluminio ed il restauro strutturale
stantial structural variation has been done during the dei ponti sospesi: il “Real Ferdinando” sul Garigliano. Restau-
execution phase: the suspension cables have been ro 146, October-December.
substituted from spiroidal wires to plate steel ele- Mazzolani, F. M. 2001. The use of aluminium in the restoration
ments similar to the shape of the original chain (Fig- of the “Real Ferdinando” bridge on the Garigliano river.
ure 12 a and b). Therefore, now the suspension sys- Festschrift zu Ehren Von Prof. Dr. Ing. Günther Valtinat.
tem looks exactly as in the old bridge with two Herausgegeben von Jürgen Priebe und Ulrike Eberwien,
couple of chains made of steel, each composed by Druck: General Anzeiger, Rhauderfehn.
four plate elements pinned at the ends. Mazzolani, F. M. 2006. Aluminium Structures in Refurbish-
Such choice seems to be very questionable from the ment: the case of the “Real Ferdinando” bridge on the
conceptual point of view, because this structural el- Garigliano river. Structural Engineering International 16 (4):
352 – 355.
ement has been rebuilt following typology and shape Mazzolani, F. M. & Mele, E. 1997. Use of aluminium alloy in
related to the past technology, but using modern retrofitting ancient suspension bridges. Proc. of the Int. Con-
technology and material, in total disagreement with ference on Composite Structures, Innsbruck, September.
the basic principles of the modern restoration theory.
Another weak point of the execution consists on the
fact that, contrary to the initial decision, the alumini-
um surface of the deck structures has been painted
with a rust-coloured coating, which hides the alu-
minium and makes the surface looking as the origi-
nal old iron. This produces a twofold falsity: artifi-
cial aging and unuseful corrosion protection, both
simulating a corroded surface. The same procedure
has been used also for the steel elements of the
bridge, i.e. chains and suspension ties. Despite these
licenses, the basic structural choice to have an alu-