Conference PaperPDF Available

Review on Long-term Behaviour of Timber-Concrete Composite Floors

Authors:

Abstract and Figures

Timber-concrete composite (TCC) beams are made up two materials, i.e. wood and concrete, which exhibit different behaviours under long-term loading. The time-dependent behaviour of TCC beam is not only affected by the long-term load but also driven by the variation of the environmental conditions such as temperature and relative humidity. In particular, the maximum deflection under service loads may govern the design requirement for medium to long span TCC beams subjected to heavy environmental conditions. For such structures, application of simplified methods adopted by different codes may lead to significant errors. Hence investigating the long-term behaviour of TCC beams subject to variable environmental condition is of great importance for designers and researchers. In this paper the research undertaken on long-term behaviour of TCC floors is critically reviewed and the recent findings are highlighted. The most important references in the literature were selected to provide more depth into the time-dependent performance of TCC structure.
Content may be subject to copyright.
From Materials to Structures: Advancement through Innovation – Samali, Attard & Song (Eds)
© 2013 Taylor & Francis Group, London, ISBN 978-0-415-63318-5
Review on long-term behaviour of timber-concrete composite floors
N. Khorsandnia, H.R. Valipour, R. Shrestha, C. Gerber & K. Crews
Centre for Built Infrastructure Research, University of Technology, Sydney, NSW,Australia
ABSTRACT: Timber-concrete composite (TCC) beams are made up two materials, i.e. wood and concrete,
which exhibit different behaviours under long-term loading. The time-dependent behaviour of TCC beam is
not only affected by the long-term load but also driven by the variation of the environmental conditions such
as temperature and relative humidity. In particular, the maximum deflection under service loads may govern
the design requirement for medium to long span TCC beams subjected to heavy environmental conditions. For
such structures, application of simplified methods adopted by different codes may lead to significant errors.
Hence investigating the long-term behaviour of TCC beams subject to variable environmental condition is of
great importance for designers and researchers. In this paper the research undertaken on long-term behaviour of
TCC floors is critically reviewed and the recent findings are highlighted. The most important references in the
literature were selected to provide more depth into the time-dependent performance ofTCC structure.
1 INTRODUCTION
Many timber girder-concrete deck bridges were built
about 65–75 years ago.Then after about a 30-year time,
there was an increase in the trend of using compos-
ite material systems and therefore, the development
of timber-concrete composite (TCC) beams has been
increasing in building and bridge structures. The TCC
beam represents a construction technique widely used
in building new structures or upgrading the strength
and stiffness of existing structures.This technique con-
sists of connecting an existing or new timber beam
with a concrete slab cast above the timber decking by
means of a connection system.
Having a composite section with concrete flange
and timber web combines the best properties of both
materials to be exploited and consequently is a good
motivation for using them. Also, timber is a renew-
able resource based material which reduces the energy
of material production and carbon dioxide emissions
compared with steel and concrete. The advantages of
these composite systems compared to floors made
solely from timber are great; with an increase of
loading capacity which reduces the deflection under
imposed loads, good acoustic performance, higher fire
resistance and less susceptible to vibrations.
Recently, extensive research has been developed to
investigate different behaviours of TCC beams and the
significance of this composite structure was proven
through them (Ceccotti & Covan 1990; Fragiacomo
et al. 2007a). The research can mainly be divided
into experimental and analytical investigations. The
experimental programs usually include some tests on
small and full scale specimens to capture the structural
behaviour under the applied force, and the analytical
studies involve numerical or finite element modelling
to predict the relevant behaviour in an efficient and
accurate method.
TCC structures include three different components,
i.e. timber, concrete and connection, which resist
not only the same forces, but also show different
behaviours during their life-time. The concrete slab
usually resists compression, while the timber usually
resists tension induced by bending, and the connection
system mostly provides a mechanism for shear trans-
fer. This mechanism resists slip between timber and
concrete in TCC structures.
2 BEHAVIOUR AND PERFORMANCE
OF TCC FLOORS
According to design codes, TCC structures have to sat-
isfy ultimate and serviceability limit states (ULS and
SLS) design requirements (CEN 2008) under short-
and long-term behaviour. In ULS the ultimate stress
under imposed loads and the maximum capacity of
the section govern the design, whereas in SLS the
maximum deflection can be the major criterion and
governs the design. In many cases, the maximum
deflection limit governs the design; however, the max-
imum stresses in timber are between 80% and 90%
of the design strength. Based on the above mentioned
behaviours, the adequacy of TCC floors have to be
investigated at least in two stages: (1) the time of load-
ing (short-term); and (2) the end of the service life,
usually 50 years (long-term).
The short-term behaviour of a TCC beam is gener-
ally well predicted by analytical formulations derived
from the solution of governing differential equations.
1053
These methods treat timber, concrete and the con-
nection behaviours as linear-elastic, assume equal
curvatures for structural elements, and neglect shear
deformations.
The long-term behaviour of these composite sys-
tems is more complicated because the long-term
behaviours of timber, concrete and connection which
are affected by creep and shrinkage/swelling, are dif-
ferent. The maximum deflection under serviceability
limit state (SLS) loads is the most influential parame-
ter on the performance of medium to long span TCC
beams subjected to severe environmental conditions
(Fragiacomo & Schänzlin 2010). Hence investigat-
ing the behaviour of such structures under long-term
loading is important.
2.1 Long-term behaviour of TCC components
The time-dependent behaviour of TCC beam is not
only affected by the long-term load but also driven by
the variation of the environmental conditions such as
temperature and relative humidity. There are also some
simplified approaches in dimensioning codes to check
the long-term behaviour of TCC structures but they
may lead to significant errors especially for long-span
beams under variable climates (Amadio et al. 2000).
All the materials employed in TCCs, i.e. concrete,
timber and connection system, display significant
time-dependency which changes both strain and stress
distribution. These temporal effects depend upon sev-
eral factors such as stress level, moisture content,
temperature and relative humidity of the environment.
Furthermore, the behaviours of TCC components are
complicated and strongly non-linear under the com-
bined effect of thermo-hydromechanical cycles and
therefore, thermal and moisture variations should be
accounted for in the design of long-term composite
beams, especially when a more accurate estimation of
deflection is needed. Since a TCC beam is an inde-
terminate system, stress redistributions can occur in
concrete, timber and connection (Mueller et al. 2008;
Yeoh et al. 2010) due to different time-dependent
properties of the materials which make the structural
problem different than a pure creep problem.
The durability and behaviour of timber in the long-
term can be influenced by the viscoelastic creep,
mechano-sorptive effect (i.e. the increment of delayed
strain due to changes of moisture content), shrinkage/
swelling and the variation ofYoung’s modulus due to
moisture content changes. When the stresses are less
than 35% of the short-term strength and the environ-
ment condition is stable, timber can be regarded as a
linear-viscoelastic material, whereas in a variable envi-
ronment, the viscoelastic behaviour of timber becomes
non-linear. Relative humidity changes cause moisture
content changes which can affect timber behaviourand
lead to fairly large deformations in timber elements.
Creep in changing moisture conditions is larger than
creep in constant wet conditions, which is also larger
than creep in constant dry conditions. So the long-term
behaviour of timber is a complex function of both the
ambient conditions and the inherent properties of the
structure, i.e. wood species, dimensions, coating, etc.
The long-term behaviour of concrete is also affected
by different phenomena such as creep, shrinkage and
thermal expansion and they all have to be taken
into account. Moreover, the long-term mechanical
behaviour of connecting system (creep of connec-
tion) has a great influence on TCC efficiency. The
slip modulus and the strength of a mechanical con-
nector do not demonstrate its creep behaviour while
this phenomenon depends mainly on the shear trans-
fer mechanism which affects local stress distribution in
timber and concrete. Connection systems show a non-
linear behaviour even for small shear forces; therefore
it is necessary to take account of this non-linearity to
evaluate the loading capacity and long-term behaviour
of the structure correctly.
Accordingly, much of the recent research has been
undertaken on the long-term behaviour of TCC beams.
The research can be mainly categorized as studies
based on codes and standard provisions, analyti-
cal investigations and experimental works which are
reviewed in the following sections.
3 CODES AND STANDARD PROVISIONS
The long-term behaviour of TCC floors are evalu-
ated in different codes and standards based on the
creep behaviour of timber, concrete and connection.
Amongst them Eurocode (CEN 2008) is the most
complete for TCC sections. Several recent studies
have been conducted based on Eurocode provisions,
some of which are reviewed in here.
The long-term deflection of a TCC beam based on
Eurocode 5 provision, which recommend to double
the creep coefficient for the connection, was evaluated
by Kavaliauskas et al. (2005). The Eurocode method
only gives initial and final deformation, while the
experimental results indicate that the predicted final
deformation is often exceeded within the first year for
medium and long span floors. This research proposed
calculating the creep of concrete and timber separately,
with concrete creep deformation calculated based on
Eurocode 2 provisions and the timber creep deforma-
tion calculated using an exponential law based on the
work of Fragiacomo & Ceccotti (2004). Results from
the proposed method showed that the initial deflec-
tion prediction was twice that predicted by Eurocode
5 and reached almost its final value over a period of
60 days. The most changes in stresses and deflections
occurred during the first 180 days and the calcu-
lated final deflection was 1.5 times that predicted by
Eurocode 5.
Schänzlin & Fragiacomo (2007) introduced two
rigorous approaches for evaluating long-term deflec-
tions. One approach used material creep coefficients
to evaluate the effect of inelastic strains such as con-
crete shrinkage. Another approach used the effective
creep coefficients of the different materials instead
of the pure material creep coefficients to calculate
1054
the difference of creep rates over time. Also concrete
shrinkage and climatic strains were transformed into
an equivalent uniformly distributed load to facilitate
using Eurocode 5 formulas for composite beams with
flexible connections.
Schänzlin & Fragiacomo (2008) also introduced
a design method which is called “effective creep
coefficients” based on the stress redistribution. More-
over, the method of determining the effective creep
coefficients and the necessity and accuracy of them
were reported. It was shown that the stress redis-
tribution affects the effective creep coefficient and
has a negligible influence on the behaviour of TCC
beams exposed to outdoor conditions with small
γ-coefficients according to the Eurocode 5; however,
it can increase the deflection up to 30% for large
γ-coefficients particularly in indoor conditions. More-
over, the bending moment of timber is also influenced
by the stress redistribution and it can lead to a 30%
underestimation by using only pure creep coefficients.
Finally, based on the parametric study, some simple
conservative formulas have been proposed to calculate
the increase in deflection and timber bending moment
due to the stress redistribution.
In another paper, Schänzlin (2008) compared the
creep coefficients of different standards and found a
quite variability of possible values. It was concluded
that the differences are not only quantitative but also
qualitative. For example, in the German standard (DIN
1052) the stress level affects the creep deformation,
whereas in Eurocode 5 the creep strain is only affected
by the service class.
4 ANALYTICAL AND NUMERICAL STUDIES
The moisture content variations markedly affect the
mechano-sorptive of timber and connection and con-
sequently the creep of timber,concrete and connection.
These phenomena influence the stress distribution and
deflection of the composite beam and therefore, the
effects of the moisture content variation on the per-
formance of TCC structures should be modelled by
rigorous numerical algorithm or finite element model.
Accurate analysis of deformations in TCC beams
requires: (1) analysis of moisture changes at each
location of the beam as a result of relative humidity
changes, (2) correct prediction of creep development
at each location as a function of moisture changes,
(3) consideration of the changing stress distribution
and displacement due to hygroexpansion and creep.
Having an exact analytical methods and simulations
can reduce the number of experimental tests which
have to be carried out to find a correct design for TCC
structures.
The developed analytical methods to date can be
basically classified to frame elements and continuum-
based models. The continuum-based models offer a
good versatility without the need for push-out tests
in order to characterize the connection properties, but
they are time consuming from computational point of
view and may not be efficient for rigorous long-term
analysis of TCC beams.
4.1 Frame FE models
A finite difference method were developed in order to
solve the viscoelastic problem forTCCs with rigid con-
nections (Mungwa & Kenmou 1993). Later, another
FE model with deformable shear connectors and dis-
placement formulation was developed byAmadio et al.
(1999) to study creep and shrinkage effects in TCC
beams. The capabilities of the proposed formulation
have been demonstrated within a simple numerical
example. A non-linear FE-program for assessment of
the long-term performance of timber beams in vari-
able surrounding conditions is described in Hanhijarvi
(2000). The computational results include long-term
deflection and time-to-failure predictions. The method
shows a good potential to predict the long-term
response of timber with promising results. Further, a
numerical procedure was developed by Amadio et al.
(2000) to evaluate the long-term behaviour of TCC
beams under a sustained load and considering seasonal
average moisture and temperature variations with a
sinusoidal law as well as connection deformability.
Recently, Fragiacomo has investigated the long-
term behaviour of TCC beams (Fragiacomo 2005,
2006; Fragiacomo & Ceccotti 2006). In these papers,
deformability of the connection system as well as rheo-
logical behaviours of concrete, timber and connection
were considered. Linear models were used to evaluate
different components of creep and the effect of mois-
ture content on the creep of timber and connection. The
proposed finite element is shown in Figure 1 which is
constituted by two parallel beams. The cross-sections
of both beams were discretised into fibres in order to
take account of different properties along the depth and
the width. Also the connection system was modelled
by linking upper and lower beams using a continuous
spring system along the beam axis.
Fragiacomo et al. (2007b) conducted another
numerical research program to analyse the TCC beams
with notched connection system to extent the exper-
imental results of 133 days to the entire service life
of the structure. The proposed FE model underesti-
mated the elastic deflection of the notched beam, but
the overall time-dependent behaviour was predicted
with reasonable accuracy. The results showed quite
large deflections (about one hundredth of the beam
length) over the entire service life; however the stress
Figure 1. The finite element used to model the composite
beam (Fragiacomo 2005).
1055
variation was not significant. Moreover, a simplified
approach developed for shored composite beams with
smeared connection has been extended to unshored
composite beams with notched connection for pre-
dicting the long-term behaviour. The results of the
analytical formulas correlate very well with the test
results, especially the deflections. The proposed ana-
lytical method can be used for the design of notched
composite beams.
4.2 Continuum-based FE models
A three dimensional model implemented in the finite
element code ABAQUS is presented by Bou Saïd,
Jullien & Ceccotti (2004) which predicts the behaviour
of TCC floors exposed to environmental conditions.
The comparison between FE results and experimental
measurements demonstrate the accuracy of the pro-
posed models and their efficiency for predicting the
behaviour of composite structures under service loads.
To (2009) focused on developing an advanced numer-
ical model of the time-dependent behaviour of the
layered wood-concrete composite with notched shear
keys by using 3D finite element models. The mod-
elling includes the diffusion, heat transfer and the
mechanical response. The main goal of the research
was to expand available constitutive laws of timber
and concrete for 3D FEM models. The results of the
verification analysis had close correlation with the
observed time-dependent behaviour of the test speci-
men for the first 123 days of the test. Fortino & Toratti
(2010) developed a moisture-stress analysis based on
a 3D FEM model by using the ABAQUS code. The
wood was modelled as an orthotropic material and
viscoelastic and mechano-sorptive behaviour for wood
was considered too.The aim of the research was to esti-
mate the levels of moisture induced stresses in timber
connections and particularly in dowel-type joints with
computational method. The method was used for the
moisture-stress analysis of a dowel-type joint under
natural relative humidity.
5 EXPERIMENTAL WORKS
Further long-term tests are needed to validate the
design guidelines and calibrate the developed exist-
ing analytical and numerical models, although they
are expensive and time consuming. The tests can
be divided to two main groups, i.e. connections and
beams, and most of the recent ones are reviewed in the
following subsections.
5.1 Connections
Fragiacomo et al. (2007a) performed some experimen-
tal tests on the “Tecnaria” shear stud connection sys-
tem for TCC beams. Some specimens constructed and
were subjected to push-out tests in the long-term under
sustained load in constant and variable environmental
relative humidity. The specimens included both nor-
mal weight and light weight concrete. Also, Mueller
et al. (2008) arranged long-term shear tests as push-out
tests at the Bauhaus-University of Weimar on three dif-
ferent types of connectors (i.e. groove, stud connector
and X-connector). The influence of variations of mois-
ture and temperature on the stiffness of the connectors
under long-term loading was investigated. In addition,
the outcomes of long-term tests under sustained load
on three other types of connections (i.e. groove and
notch with coach screw and tooth metal plate with per-
forated holes) have been reported byYeoh et al. (2010).
The tests were done over a period of almost 1.5 years
in an unheated, poorly controlled indoor condition.
Key parameters such as relativehumidity, temperature,
moisture content and the relative slip of the connec-
tions were measured under environmental condition of
service class 3 according to Eurocode 5.
5.2 Beams
Different experimental programs have been conducted
on TCC beams to investigate the load-deflection
behaviour of the structure over time. To evaluate the
effect of changing environmental conditions, such as
relative humidity and temperature, various ambient
conditions have been employed. The ambient con-
ditions can be generally categorized to controlled
indoor (Bonamini et al. 1990; Ceccotti & Covan 1990;
Yeoh et al. 2010), uncontrolled indoor (Fragiacomo
et al. 2007b), sheltered outdoor (Bou Saïd et al. 2004;
Kenel & Meierhofer 1998; To 2009) and unsheltered
outdoor (Capretti & Ceccotti 1996; Ceccotti et al.
2007; Schänzlin 2008). In terms of test duration, they
can be also classified to less than one year (Bonamini
et al. 1990; Fragiacomo et al. 2007b;To 2009), between
one to two years (Bou Saïd et al. 2004; Yeoh et al.
2010) and about five years (Capretti & Ceccotti 1996;
Ceccotti et al. 2007; Kenel & Meierhofer 1998).
Furthermore, different length and section dimensions
have been used in the TCC beams. With regard to
the beam length, they can be divided into short span
(less than 3 m) (Bonamini et al. 1990; Ceccotti &
Covan 1990;To 2009), medium span (between 3 m to
6 m) (Capretti & Ceccotti 1996; Ceccotti et al. 2007;
Fragiacomo et al. 2007b; Kenel & Meierhofer 1998)
and long span (more than 6 m) (Bou Saïd et al.
2004; Yeoh et al. 2010). In most of the tests, differ-
ent parameters such as, mid-span deflection, relative
slip between timber and concrete, moisture content,
relative humidity and temperature have been recorded
over time.
6 RECENT RESEARCH OUTCOMES
Some of the major outcomes of the reviewed papers
can be summarized as follows.
The variations of environmental conditions influ-
ence the behaviour of TCC beams, the elastic
1056
modulus of timber and the mechano-sorptive creep
of both timber and connection system. Such stresses
and deflection, may become quite significant, espe-
cially for composite beams with stiff connec-
tions and medium to long span length (Amadio
et al. 2000; Ceccotti et al. 2007; Fragiacomo &
Ceccotti 2006; Fragiacomo & Schänzlin 2010;
Mueller et al. 2008).
The mechano-sorptive creep is not only affected
by the level of moisture content but also can be
influenced significantly by the number of changing
moisture cycles especially for longer span beams
(To 2009;Yeoh et al. 2010).
The maximum moisture variation takes place on the
surface exposed to the environment, whilst the min-
imum occurs in the centre of the timber section.
Therefore, the width of the timber beam plays a
significant role on the amplitude of the moisture
content variation (Fragiacomo & Schänzlin 2010;
To 2009). Further, using the average moisture con-
tent between the outer surface and the centre of the
timber section and a constant temperature over the
cross section of timber and concrete can be adequate
for design purposes (Fragiacomo & Ceccotti 2006).
The daily and weekly variations of the environmen-
tal relative humidity history do not significantly
affect the trend in time of the average moisture
content and therefore, considering weekly or larger
cycles in long-term behaviour of timber beams is
adequate (Fragiacomo & Schänzlin 2010; Schän-
zlin 2008). It was also concluded that the influence
of initial moisture content on the time dependent
stiffness of the beam is minor (Schänzlin 2008).
The time-dependent response of the tested speci-
mens is sensitive to the change of temperature and
the effect of thermal exposure should be included
in load combinations used in any design code (To
2009).
The concrete shrinkage, connection creep, mois-
ture content and temperature variations should be
considered in the long-term analysis and design of
composite beams, especially when a more accurate
evaluation of the deflection is required (Fragiacomo
2006; Fragiacomo & Schänzlin 2010; Mueller et al.
2008).
The connection system shows mechano-sorptive
creep due to the hygroscopic behaviour of timber
around the connector. The amount of deformation
is negligible for the relative humidity cycles with
periods less than one week and maximum ampli-
tude of 40% (Fragiacomo et al. 2007a). Increasing
the stiffness of the connector would increase the
total stiffness of the structure, and consequently the
deflection can be reduced (Mueller et al. 2008). It is
also observed that the creep of connection is bigger
than creep of timber (Capretti & Ceccotti 1996).
The variation of the stresses during the service life is
not as significant as for the deflection, i.e. the long-
term behaviour of TCC beams mostly influence the
serviceability limit state, rather than internal forces
(Fragiacomo et al. 2007b).
The use of light weight concrete instead of nor-
mal weight concrete does not significantly affect the
performance of the TCC beams neither in the long-
term, nor in the collapse tests (Fragiacomo et al.
2007a).
An increase in deflection was mainly observed dur-
ing the first two years of the long-term test, while the
slip rose during the whole testing period (Ceccotti
et al. 2007)
A TCC beam in outdoor conditions should be clas-
sified as service class 3 based on Eurocode 5
provisions regardless of the maximum value of the
timber moisture content (Ceccotti et al. 2007).
The construction type (shored or unshored) does not
significantly affect the structural performance of
the composite system (Fragiacomo et al. 2007b),
but it can influence the initial deflection.
If the deflection is required to be limited for service-
ability performance, some methods such as, prop-
ping, pre-cambering the beam, using low shrinkage
concrete and precasting the concrete to let it shrink
before construction are recommended (Fragiacomo
et al. 2007b; Yeoh et al. 2010).
7 CONCLUSIONS
The long-term behaviour of timber-concrete compos-
ite structures is mainly governed by the maximum
deflection of the beam under serviceability limit state
(SLS) loads because of time dependent behaviour of
TCC components (i.e. timber, concrete and connec-
tion) in variable ambient conditions. This effect can
be the most severe design parameter for medium to
long span TCC beams subjected to heavy environmen-
tal conditions. The behaviours of TCC components
are complicated and strongly non-linear under thermo-
hydromechanical effects. Different experimental pro-
grams have been conducted to date to investigate the
long-term behaviour of TCC connections and beams,
but they are expensive to run and time consuming.
Therefore, having an accurate analytical simulation
can reduce the number of experimental tests which
have to be carried out to find a correct design for
TCC structures. The effects of the environmental varia-
tions on the performance of TCC structures should be
modelled by rigorous numerical algorithm or finite
element model to save the cost and time associated
with testing.
REFERENCES
Amadio, C., Ceccotti, A., Di Marco, R. & Fragiacomo, M.
2000, ‘Numerical evaluation of long-term behaviour of
timber-concrete composite beams’, paper presented to
the 6th WCTE 2000, Whistler Resort, British Columbia,
Canada.
Amadio, C., Di Marco, R. & Fragiacomo, M. 1999, ‘A
linear finite element model to study creep and shrink-
age effects in a timber-concrete composite beam with
deformable connections’, paper presented to the 1st
1057
International Rilem Symposium on Timber Engineering,
STOCKHOLM, SWEDEN.
Bonamini, G., Uzielli, L. & Ceccotti, A. 1990, ‘Short- and
long-term experimental tests on antique larch and oak
wood-concrete composite elements’, paper presented to
the C.T.E. Conference, Bologna.
Bou Saïd, E., Jullien, J.F. & Ceccotti, A. 2004, ‘Long term
modelling of timber-concrete composite structures in vari-
able climates’, paper presented to the 8th WCTE 2004,
Lahti, Finland.
Capretti, S. & Ceccotti, A. 1996, ‘Service behaviour of
timber-concrete composite beams: A 5-year monitoring
and testing experience’, paper presented to the Interna-
tional Wood Engineering Conference, New Orleans, USA.
Ceccotti, A. & Covan, C. 1990, ‘Behaviour of timber and
concrete composite load-bearing structures’, paper pre-
sented to the The 1990 IUFRO S5.02 Timber Engineering,
Saint John, New Brunswick, Canada.
Ceccotti, A., Fragiacomo, M. & Giordano, S. 2007, ‘Long-
term and collapse tests on a timber-concrete composite
beam with glued-in connection’, Materials and Struc-
tures, vol. 40, no. 1, pp. 15–25.
CEN 2008, Eurocode 5: Design of timber structures– Part 1 –
1: General – Common rules and rules for buildings.
EN 1995-1-1:2004+A1, BSI, European Committee for
Standardization, Brussels, Belgium.
Fortino, S. & Toratti, T. 2010, ‘A three-dimensional moisture-
stress FEM analysis for timber structures’, paper pre-
sented to the 11th WCTE 2010, Riva del Garda, Trentino,
Italy.
Fragiacomo, M. 2005, ‘A f inite element model for long-term
analysis of timber-concrete composite beams’, Structural
Engineering and Mechanics, vol. 20, no. 2, pp. 173–90.
Fragiacomo, M. 2006, ‘Long-Term Behavior of Timber-
Concrete Composite Beams. II: Numerical Analysis and
Simplified Evaluation’, Journal of Structural Engineer-
ing, vol. 132, no. 1, pp. 23–33.
Fragiacomo, M., Amadio, C. & Macorini, L. 2007a, ‘Short-
and long-term performance of the “Tecnaria” stud con-
nector for timber-concrete composite beams’, Materials
and Structures, vol. 40, no. 10, pp. 1013–26.
Fragiacomo, M. & Ceccotti, A. 2004, ‘A simplified approach
for long-term evaluation of timber-concrete composite
beams’, paper presented to the 8th WCTE 2004, Lahti,
Finland.
Fragiacomo, M. & Ceccotti, A. 2006, ‘Long-Term Behavior
of Timber-Concrete Composite Beams. I: Finite Ele-
ment Modeling and Validation’, Journal of Structural
Engineering, vol. 132, no. 1, pp. 13–22.
Fragiacomo, M., Gutkowski, R.M., Balogh, J. & Fast, R.S.
2007b, ’Long-Term Behavior of Wood-Concrete Com-
posite Floor/Deck Systems with Shear Key Connection
Detail’, Journal of Structural Engineering, vol.133, no. 9,
pp. 1307–15.
Fragiacomo, M. & Schänzlin, J. 2010, ‘The effect of moisture
and temperature variations on timber-concrete composite
beams’, paper presented to the 11th WCTE 2010, Riva del
Garda, Trentino, Italy.
Hanhijärvi, A. 2000, ‘Computational method for predicting
the long-term performance of timber beams in vari-
able climates’, Materials and Structures, vol. 33, no. 2,
pp. 127–34.
Kavaliauskas, S., Kvedaras, A.K. & Gurksnys, K. 2005,
‘Evaluation of long-term behaviour of composite timber-
concrete structures according to EC’, Technological and
Economic Development of Economy, vol. 11, no. 4,
pp. 292–6.
Kenel, A. & Meierhofer, U. 1998, Long-term performance
of timber-concrete composite structural elements, Repor t.
No. 115/39, EMPA, Dübendorf, Switzerland.
Mueller, J., Haedicke, W., Simon, A. & Rautenstrauch, K.
2008, ‘Long-term performance of hybrid timber bridges –
experimental and numerical investigations’, paper pre-
sented to the 10th WCTE 2008, Miyazak, JAPAN.
Mungwa, M. & Kenmou, D. 1993, ‘Instantaneous and
time-dependent analysis of composite wood-concrete
cross-sections using Dischinger’s equations of state:
Part 1-Instantaneous analysis’, Materials and Structures,
vol. 26, no. 2, pp. 98–102.
Schänzlin, J. 2008, ‘Comparison of existing long term
deformations with models’, paper presented to the
10th WCTE 2008,Miyazak, JAPAN.
Schänzlin, J. & Fragiacomo, M. 2007, Extension of EC5
Annex B formulas for the design of timber-concrete
composite structures, In: Meeting forty of the Working
Commission W18- Timber Structures, CIB, International
Council for Research and Innovation, Bled (Slovenia).
Schänzlin, J. & Fragiacomo, M. 2008, ‘Modelling and design
of timber-concrete-composite structures in the long-term’,
paper presented to the 10thWCTE 2008, Miyazak, JAPAN.
To, L.G. 2009, ‘3D finite element modelling of time-
dependent behavior of wood-concrete composite beams’,
Ph.D. thesis, Colorado State University, Fort Collins,
Colorado.
Yeoh, D., Fragiacomo, M. & Deam, B. 2010, ‘Long-term
behaviour of LVL-concrete composite connections and
beams under sustained load’, Materials and Structures,
unpublished.
1058
... The review presented in the following deals with the research studies available in literature regarding hybrid timber-based buildings and building components, where timber is combined with concrete and steel. Other review articles on the topic have been found in literature, however they focus only on some aspects of hybrid timber-based structures: mechanical performance of TCC structural components [30,45], fire resistance of TCC members [46], seismic behaviour of hybrid timber-based buildings [47]. The aim of the present article is to give a comprehensive overview of the state of the art of hybrid timber-based buildings and building components, in order to establish a background for improving existing solutions and proposing new timber-based hybrid systems and components. ...
... In case of fire, the timber cross-section is reduced due to charring and the impact of high temperatures can lead to rapid loss of the connection stiffness. Currently, Eurocode 5 does not include design procedures for the calculation of the fire resistance for TCC and STC components, however they are currently under development and will be included in the new generation of the Eurocode 5 [45]. ...
Article
The construction of timber buildings has increased in recent years, thanks to the excellent properties of the material. To achieve improved behaviour in terms of mechanical properties, energy and acoustic performance, fire resistance and durability, timber structures are sometimes integrated with other materials, such as concrete and steel, resulting in hybrid timber-based structures. This paper presents a literature review on hybrid timber-based structures, summarizing the state of the art of hybrid timber-based structures constructed to date and examining the main research contributions. The aim is to establish a background for improving existing solutions or proposing new hybrid timber-based systems and components.
... The creep and shrinkage coefficients of concrete and timber are different; accordingly, the long-term behaviour of the CLT panel, concrete slab, and connections also vary [51]. Excessive environmental fluctuation causes a large variation in concrete and CLT's creep, shrinkage, and thermal properties [99]. The moisture level and repetition of the moisture cycle influence the mechano-sorptive creep of TCC panels, which increases with the span of the panel. ...
Article
Composite construction elements are gaining extensive attention due to their high performance and reliability. Timber–concrete composite is one of the well-known engineered products of global interest. The cross-laminated timber (CLT) is a plate-like quasi-rigid composite that is usually composed of an uneven number of layers of solid timber board. The timber boards are generally placed side-by-side and arranged crosswise in CLT. The anisotropy in single timber boards is reliably adjusted, and its out-of-plane load bearing capacity is improved. The CLT members show low bending strength and resistance to global stability. The CLT–concrete composite has developed and showed approximately 3–5 times higher strength capacity than the conventional timber or concrete structures. This study aims to review the current practice and available guidelines for the CLT–concrete floor system. The basic required information on the design and construction and performance of the aforementioned system is described based on the current practices. This review can guide prospective researchers and users regarding the application of the CLT–concrete composite. The CLT–concrete elements need to be designed and constructed following the current code of practice and studies guidelines to achieve optimal performance. The design, construction, and performance of the CLT-concrete system are dependent on timber properties, connection systems, and slab details. This composite floor system possesses good performance under various loading conditions if adequately designed and constructed.
... A detailed overview of the long-term experimental and numerical research on TCC connections and TCC structural behaviour is given in [83] and [167]. The planning and realisation of experimental long-term tests is complex, space-consuming, and therefore expensive. ...
Thesis
The combined use of timber and concrete for slabs in buildings offers a range of advantages compared to timber-only slabs: increased stiffness and load-bearing capacity, improved sound insulation and vibration behaviour, and improved fire safety. Other substantial advantages are low self-weight, fast production, and significantly lower environmental impact compared to slabs made entirely of reinforced concrete. Timber-concrete composite (TCC) slabs have been used in Switzerland for several decades. The connection between timber and concrete is of significant importance for the load-bearing behaviour of the slab. In fact, there are various connection systems on the market which have different characteristics regarding connection stiffness, load-bearing capacity, and ductility. For example, notched connections achieve an almost rigid composite action and have a high load-bearing capacity. They are supplemented by screws or dowels, supposed to carry vertical tensile forces and prevent uplift of the concrete layer. However, the material and installation costs for the screws and milling process for the notches represent an economical disadvantage. This thesis therefore presents a novel connection system called micro-notches enabling a composite slab of timber and concrete only, without additional metal connectors. The concept of micro-notches was developed in an extensive experimental program, supplemented by analytical and numerical investigations. In a first step, the connection with micronotches was tested in local shear tests in a symmetrical push-out test setup. The objectives of these investigations were to determine the connection properties of different micro-notch configurations and to find an optimum configuration. This optimum configuration was then tested in global 4-point bending tests to assess the realistic structural behaviour of the novel connection system in TCC slabs. The tested beams showed a very stiff, linear elastic load-bearing behaviour with brittle ultimate failure in the timber cross-section at mid-span. The failure mode observed in the micro-notches was for both local and global tests a shear failure of the concrete notches and, in some places, also a shear failure of the timber notches. Two models were developed to assess the global load-bearing behaviour, including an analytical model based on the gamma-method and a numerical model based on a strut-and-tie approach. The global load-bearing behaviour was then compared with the local load-bearing behaviour observed in the local shear tests. With these results, the connection properties of the micro-notches were adjusted. To assess the influence of rheological phenomena in timber, concrete, and the connection, a long-term test series was installed and first results are presented. As a further field of application for TCC slabs, a two-span beam test series was performed to evaluate the positive effect of continuous systems on the deflections. The cracking of concrete over the middle support greatly influences the beam stiffness and needs to be considered in the calculation models. The tests showed that micro-notches are also functioning in cracked concrete and that the multi-span system reduces deflections. The findings of the experimental investigations and the developed models described above are summarised in design and production recommendations.
Article
This paper describes an analytical procedure for designing timber-concrete composites (TCC) subjected to boundary conditions other than simply supported. Currently available investigations of TCCs are mainly focused on simply supported slabs, as it is a typical configuration for timber buildings. However, in other structural applications, and remarkably for reinforced concrete buildings, the boundary conditions of the TCC slabs are not likely to be simply supported. Such distinct boundary conditions can significantly reduce the cross section height, mid-span deflection and self weight of the structure, the last one being crucial in seismic regions. The proposed procedure is derived from two simplified methods available in the literature, one general in its nature while the other being valid for simply supported beams. The short-term analytical model was compared against finite element models (FEM) and to the only experimental investigation on partially restrained TCCs available in the literature, while the long-term analytical model was compared only against FEM. At the end of the investigation, a full-scale continuous TCC beam was tested in the serviceability range, to compare with the prediction of the proposed analytical model. The model underestimated the mid-span deflection at 4.5 kN by 13%, concluding that the proposed simplified procedure is valid for boundary conditions other than simply supported. Further experimental campaigns are needed in the future to assess the versatility of the model in a wider range of boundary conditions, including short-term and long-term tests, which should enhance the applicability of TCC slabs in structures different from timber buildings and bridges.
Article
Full-text available
The reinforcement of a timber floor with a concrete layer properly connected to produce a composite system represents not only an improvement in the stiffness but also a foreseeable change in the modal frequencies of the floor. In order to assess that change, a typical timber floor, which was subsequently reinforced with a concrete layer to produce a composite, was tested at both stages in the research reported in this paper. Additionally, composite beams taken from the original floor were also tested. The results showed a significant decrease (44%) in the fundamental frequency after the reinforcement. In order to obtain the mode shapes and frequencies of the floor, finite-element (FE) models of the floors and beams were created. The FE models accurately matched the experimental results. Having validated the model, a parametric study was carried out to compare the results provided by the proposed FE model with a numerical model from the bibliography and to understand the influence of the most relevant structural parameters on the fundamental frequency of the composite floor.
Article
Full-text available
The objective of this work is the investigation of long‐term behaviour of composite timber‐concrete beams. Both ultimate and serviceability limit states have to be satisfied according to modern design codes, such as Eurocode 5. Because of the rheological phenomena such as creep, time‐dependent (viscoelastic) creep, mechano‐sorptive (moisture change) creep and shrinkage/swelling that occur in the component materials, stress and strain distribution changes in time. The rheological phenomena have to be taken into account for long term verifications. This paper presents the review of the evaluation of long term behaviour of composite timber‐concrete beams according to European design codes EC5 and EC2. Kompozitinių medienos-betono perdangų sijų ilgalaikės elgsenos vertinimas pagal euronormas Santrauka Straipsnyje atlikta kompozitinių medienos ir betono perdangų skaičiavimo pagal Euronormas analizė, pateikti jos ilgalaikės elgsenos vertinimo aspektai. Atlikti vienos tokio tipo perdangų skaičiavimai pagal EC5 pateiktas formules ‐ efektyvaus modulio metodą (Efective Modul Method), taikant konkrečius medienos ir betono medžiagų valkšnumo modelius. Reikšminiai žodžiai: Medienos‐betono kompozitas, ilgalaikė elgsena, valkšnumo modeliai, traukumas, ilgalaikiškumo modeliavimas First Published Online: 21 Oct 2010
Article
Full-text available
In the past years, hybrid constructions in form of timber-concrete composites became more interesting for bridge constructions. On that account long-time shear tests as push-out-tests were arranged at the Bauhaus-University of Weimar. The results of these long-time shear tests, which were accomplished at three different types of connectors, are presented in this paper. An important outcome of the analysis of the shear tests is a suggestion for the creep coefficients of the different connector types. In addition the different time and climatic dependent behaviour of timber and concrete has got a great influence to the load bearing capacity and serviceability of timber-concrete composite bridges. Using a numerical parametrical study an investigation of these effects was possible.
Article
Full-text available
Shrinkage and swelling of timber in the parallel to the grain direction are usually ignored in normal design as they are significantly lower than the values in the perpendicular to the grain direction. In timber-concrete composite structures, however, the shrinkage/swelling of the timber beam in the parallel to grain direction cannot freely occur due to the restraint provided by the concrete slab through the connection system. As a consequence of that, self-equilibrated stresses (eigenstresses) and additional deformations such as deflection will be induced by moisture content variations in the composite beam. The paper discusses the significance of moisture and, more in general, environmental variations including temperature on the behaviour of timber-concrete composite structures, and provides a simplified design approach to consider such effects in design.
Article
Full-text available
In this paper a numerical procedure to evaluate the long-term behaviour of timber-concrete composite beams based on a finite element method is introduced. This procedure allows to take into account in a rigorous way connection deformability, creep of components, mechano-sorptive effect of timber and inelastic strains of timber and concrete. Generic creep functions, temperature and moisture histories may be adopted. Based on some first results, the importance of these aspects on the global structural behaviour is shown. Changes of environmental conditions seem to play a very important role. The proposed procedure may then represent an useful tool to better understand the complex mechanisms of interaction between concrete slab and timber beam during time due to viscous effects in timber and concrete.
Article
Full-text available
In the paper, a simplified approach to evaluate the long-term response of timber-concrete composite beams is proposed. The effects due to dead and live load, concrete shrinkage, yearly and daily environmental variations are superimposed. Sustained load conditions, such as loads and concrete shrinkage, causes elastic, creep, and mechano-sorptive creep strains. These phenomena, which increase the deflection and modify the stress distribution in time, are considered by using the "Effective Modulus Method". The inelastic strains due to environmental variations of relative humidity and temperature are calculated by using the elastic solution for composite beams with flexible connection. The proposed approach leads to results close to those obtained by using more refined numerical programs. For the simplicity and good approximation, it is fit to be employed for practical design of composite beams, especially when it is important a detailed evaluation of the long-term structural behaviour.
Article
Full-text available
The objective of this work is the investigation of long-term behaviour of composite timber-concrete beams. Both ultimate and serviceability limit states have to be satisfied according to modern design codes, such as Eurocode 5. Because of the rheological phenomena such as creep, time-dependent (viscoelastic) creep, mechano-sorptive (moisture change) creep and shrinkage/swelling that occur in the component materials, stress and strain distribution changes in time. The rheological phenomena have to be taken into account for long term verifications. This paper presents the review of the evaluation of long term behaviour of composite timber-concrete beams according to European design codes EC5 and EC2.
Article
For the description of the long term behavior several different rheological models have been developed, considering normal creep as well as mechano-sorptive creep. However for the application the question arises which model should be used, since especially differences in a prediction of the creep strain after more than 20 years appear. Since tests can hardly be performed over such a long period, the deflection of beams in existing buildings is compared to the values given in the standards as well as to the rheological models.
Article
The paper presents a finite element model for studying timber-concrete composite beams under long-term loading. Both deformability of connection system and rheological behaviour of concrete, timber and connection are fully considered. The creep of component materials and the influence of moisture content on the creep of timber and connection, the so-called "mechano-sorptive" effect, are evaluated by means of accurate linear models. The solution is obtained by applying an effective step-by-step procedure in time, which does not require storing the whole stress history in some points in order to account for the creep behaviour. Hence the proposed method is suitable for analyses of composite beams subjected to complex loading and thermo-hygrometric histories. The possibility to accurately predict the long-term response is then shown by comparing numerical and experimental results for different tests.
Article
The paper investigates the long-term behavior of wood-concrete composite beams with notched connection detail. The experimental program comprised the characterization of the component materials (wood, concrete, and connection detail) and long-term tests on beam specimens. The beam specimens were monitored during the construction process, and for an overall period of 133 days after the application of the service load. The experimental results have then been extended to the entire service life of the structure using a one-dimensional finite-element model. It was found that the increase in moisture content due to the bleeding of the fresh concrete is not an issue for the durability of the wood deck, and the type of construction (shored or unshored) does not significantly affect the structural performance. The rheological phenomena experienced by the component materials lead to quite large deflections over the entire service life, whereas the variation in stress is not significant. If the limitation of the deflection is required for serviceability considerations, the use of concrete with reduced shrinkage and the precambering of the wood deck are to be recommended. A simplified approach based on closed form solutions for composite beams with smeared flexible connectors is finally proposed for the prediction of the long-term behavior.