Construction of Brazils first cable stayed Stress-laminated timber

footbridge
Everaldo PLETZ
Prof. Dr. Civil Engineer
State University of Londrina
Londrina - Brazil pletz@uel.br
Francisco Antonio Rocco LAHR
Prof. Dr. Civil Engineer
State University of So Paulo
So Carlos - Brazil frocco@sc.usp.br
Carlito CALIL Jr
Prof. Dr. Civil Engineer
State University of So Paulo
So Carlos - Brazil calil@sc.usp.br
Antonio Alves DIAS
Prof. Dr. Civil Engineer
State University of So Paulo
So Carlos - Brazil dias@sc.usp.br
Summary
This paper describes the construction of the first cable-stayed timber footbridge in Brazil. The
footbridge modular deck is fabricated from seven interconnected curved stress- laminated timber
plates. The pylon is made from a single log of eucaliptus citriodora, standing alone over a spatial
hinge. Pinus taeda and eucalyptus citriodora were the chosen lumbers, because they met the
sustainability requirements. The footbridge may be described as a curved bridge, supported by 12
steel bars working as stays, with an overall length of 35 meters and a walkway with of 2 meters.
The purpose of this paper is to describe the design and the construction process of this unique
structure in Brazil. Constructing this footbridge confirmed the technical and economical viability of
this new technology in Brazil.
Keywords : timber, cable-stayed footbridge, stress-laminated timber, design, construction
1- Introduction
The possibility of building a cable stayed timber curved footbridge, with a stress laminated deck
was then faced as a good contribution to this process of convincing brazilians to review their old
positions against timber construction. The first brazilian cable-stayed timber footbridge is a 2
meters wide footbridge made of pinus taeda, with a stress laminated curved deck, also by the first
time adopted in Brazil, spanning 34 meters from the Timber Structures and Wood Laboratory
building to the Strucutural Engineering Department building, at So Carlos Engineering School
which belongs to So Paulo State University in Brazil. The designing and building experience is
described in this paper.
2. Structural System and Design Parameters
A set of 37 treated 50 x 200x5200 mm pinus taeda laminations were used in the stress laminated
plates. Because of the laminationss length, no butt joints occured in the plates. At the the edges of
the plates two extra eucalyptus citriodora laminationss were adopted, according to RITTER & LEE
(1996) recommendation. A thirteen meters long treated log of eucalyptus citriodora, with a 450 mm
top diameter and a 550 mm basal diameter was used as pylon. The twelve stays are the same type of
the steel bar used to stress the SLT decks, Dywidag (ST 85-105) bars of 15 mm in diameter. The
tower two back-stays are the same steel bar, with 32 mm diameter. Connections and plates for
stress distribution are of SAE 1020 steel. All metalic devices are hot dipped galvanized.
3. Design
The brazilian code loading requirements for pedestrian bridges is 5 kPa. Although cable-stayed
footbridge bridges are usually flexibel structures, this particular footbridge is not sensitive to wind
action because of its arched shape in plan. Effects of temperature variation were also not
considered. On site measurements of these actions proved those effects to be negligible. Cable
fatigue was not investigated because it does not govern the design of footbriges like these one, used
by small groups of students, similar to the footbridge designed by CARTER e FAYERS (1994).
Fig. 1 Plan and View of the Footbridge
3.1. Structural analysis
A preliminary linear three dimensional analysis was carried out using the Ansys 5.4 FEM code,
considering modal characteristics evaluated with the help of empirical formulas. In order to
investigate the stability of the footbridge, the following loading combinations were also considered:
the two extreme spans loaded with the three central ones unloaded, and the three central ones
loaded with the four extreme ones unloaded. The low level of stresses presented by the footbridge
suggested that only a geometric non-linear analysis shoulde be carried out. Once undertaken, it
revealed that just the stay sag could interfere. Ernsts formula was used to model the stay sag.
Modal analysis is an indispensable resource for assessing the vibration serviceability of footbridges.
Carrying out modal analysis relies upon an
estimate of modal characterisitics of the
structure. Empirical formulas were used to
evaluate those properties and a further
experimental modal analysis was undertaken
to to verify the validity of the assumed values.
3.2. Sizing
Eurocode 5 part 2 establishes curves for
acceptability to vibration serviceability which
were adopted in the footbridge design. It was
the critical design condition. The height of
deck section, the bar diameter of the stays
were determined according to this critical
condition and to market availability. The 15
mm diameter bar met these requirements. To
E
L
E
V
A

O
establish the log tower dimensions, avoiding
second order effects was a main concern
because in this particular footbridge the tower
has a crucial role on the footbridge
performance.
3.3. Detailing
Easinees of replacement, minimizing
excentricities and stress concentration, were
the main concerns in connection detailing,
the most challenging part of the design.
Shrinkage and shape distortions due to
moisture content variation were also taken
into account. Timber and metallic parts were
both designed to ensure adequate durability
during the intended life. All steel work were
hot dipped galvanized and the timber, aside
from the fact of being treated, had two coats
of stain. Simplifying likely maintenance
procedures and reducing their costs were also
important aspects observed during its design.
Figure 2- Cable-stayed timber Footbridge
4. Construction
Fig.3 - Fabrication and Erection
The fabrication of members in small workshop was the chosen procedure. The deck was divided
into seven identical plates, named modules, which were fabricated in a small workshop. To inter-
connect these plates three meters long steel C beams with transversal welded plates were attached
to the ends of the stress laminated plates. This designed inter-connection system helps keeping the
deck curvature and proved to be very pratical in the assembling process. The steel C beams had to
be longer than the deck width, because the stays linked to their ends allowed them not to interfere
in the pedestrian walking area. The uneven piano-key surface at the plates ends had to be
eliminated to make possible the steel C beam attaching. The same non-uniformity of lamination
dimensions made deck width also uneven, requiring extra short laminations to be added at the end
of the plates, to allow perfect attachment to the steel C beam device, according to PLETZ (2003).
Supported horizontally by a very stiff steel beam, the module should reach a 10 cm sag with the
application of a concentraded force aplied at middle of the concave edge. The plates were
transversely pre-stressed till a 0,7 Mpa compressive tension level was developed. The resstressing
procedure indicated by to RITTER & LEE (1996) was adopted. The stressing bars were 70 cm
apart. The laminations number and dimensions of knots far exceeded the usual limits. Even
though, they were used and the transversal pre-stressing system showed one of its greatest
advantage: the defects cannot affect exclusively the plate performance, they just contribute to
reduce the global stiffness and strength. After the three concrete cassions foundation were ready, the
log of eucaliptus was lifted and supported by two back stays, made of Dywidag bars of 32 mm
diameter. After the fabrication of all modules, they were lifted on metallic scaffolding, which
allowed to apply a pre-camber to the footbridge. Children safety concerns led to the adopted hand
rail ressembling spinal shape shown in Fig. 1. To complete the footbridge two stain coats were
applied. The deck loss of curvature ( which is negligible after one year) results from the sum of the
reducing effects of the SLT plates inter-connection steel devices and of the plastic deformation of
kept curved straight laminations, with the growing effects of stressing bars tension level reduction.
This footbridge was constructed by a small building company, with the support of a small
metallurgic workshop. This footbridge costed just US$20.000. The critical paths in the time-table
of construction corresponded to substructure completion and all steel devices fabrication and
galvanization. A period of 50 days is enough for the plate fabrication and erection of a footbridge.
5. Conclusions
1. The design and construction of this footbridge proved the technical and economical viability
of the cable-stayed system and of the stress-laminated deck for construction in Brazil.
2. The use of under-graded timber, proved that the stress-laminated timber is a technology that
can meet the sustainability and technical requirements.
3. Construction of decks made only of SLT plates, simplifies building procedures and reduces
greatly the number of connections, well known points of stress concetration
4. The plate curving process described can be sucessfully applied to SLT plates.
5. Vibration serviceability proved the critical design condition.
6. Acknowledgements
This study was been supported by FAPESP Fundao de Amparo Pesquisa do Estado de So
Paulo. The authors wish to thank prof. Ernst Gheri for the valuable suggestions.
7. References
CARTER, D. J.; FAYERS,S. W. (1994) Design of a Cable-Stayed Stress-Laminated Timber
Pedestrian Bridge over Redbank Creek- in Pacific Timber Engineering Conference - Australia
EUROPEAN NORM (1995). EN 1995-1-1 Part 2 Design of Timber Bridges -
PLETZ,E (2003). Passarela Estaiada com Tabuleiro de Madeira Laminada Protendida em Mdulos
Curvos Tese de Doutorado - Escola de Engenharia de So Carlos Universidade de So Paulo.
RITTER, M.A. & LEE,P.H. (1996). Recommended Construction Practices for Stress Laminated
Wood Bridges Decks. International Wood Engineering Conference, New Orleans, EUA