See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/45313963

Proračun i modeliranje spojeva rešetkastih čeličnih nosača MKE metodom

Article · January 2010

Source: OAI

CITATIONS READS

0 440

3 authors, including:

Damir Markulak
University of Osijek
33 PUBLICATIONS   123 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Seismic Design of Frame Structures with Infill Panels (149-1492966-1536) View project

Mechanical properties of available structural aluminium alloys View project

All content following this page was uploaded by Damir Markulak on 03 June 2016.

The user has requested enhancement of the downloaded file.

Strojarstvo 52 (2) 125-135 (2010) I. RADIĆ et. al., Design
�������������������������������
and FEM Modelling...����
125

CODEN STJSAO ISSN 0562-1887

ZX470/1437 UDK 624.072.7:624.078.45:519.6

Design and FEM Modelling of Steel Truss Girder Joints

Ivan RADIĆ, Damir MARKULAK and Original scientific paper

Miljana MIKOLIN This paper critically presents the codified design of joints in steel truss
girders, according to the latest european norms [1-2] , as contemporary
Građevinski fakultet Osijek Sveučilišta J.J. norms in which this issue is included and processed in detail for the first
Strossmayera u Osijeku (Faculty of Civil time. Rules in [2] are based on simplified analytical models in combination
Engineering, J.J. Strossmayer University of with experimental research, so the regulations consist of semi-empirical
Osijek), Crkvena 21, HR - 31000 Osijek calculation formulae which are valid in limited conditions. Thus, in
Republic of Croatia engineering practice it is required to take care of parameters which affect
the global or local behaviour of truss girder such as secondary bending
radic@gfos.hr moments, bending moments resulting from eccentric member joints, and
joint deformability, so this paper gives distinct attention to these effects.
Typical truss girder behaviour modelling modes using finite element
method (FEM) are illustratred so to predict their values, and take them
in the final estimate if needed. The steel truss girder made of hollow
rectangular sections with welded joints is reviewed in detail and numerically
exemplified. The accuracies are compared and recommendations are given
for the application of each model.
Proračun i modeliranje spojeva rešetkastih čeličnih nosača MKE
metodom
Izvornoznanstveni članak
Keywords U radu se kritički sagledava proračun spojeva rešetkastih čeličnih nosača
prema novim europskim normama [1-2], kao suvremenim normama u
Finite element method
kojima se ova problematika prvi put detaljnije kodificira. Pravila u [2]
Joints se baziraju na pojednostavljenim analitičkim modelima u kombinaciji s
Rectangular hollow sections eksperimentalnim ispitivanjima, pa se propisi zapravo sastoje od semi-
Truss girder empirijskih proračunskih izraza koji vrijede u dosta ograničenim uvjetima.
Ključne riječi Stoga je u inženjerskoj praksi potrebno voditi računa o parametrima koji
Metoda konačnih elemenata utječu na globalno i lokalno ponašanje rešetkastih nosača, kao što su npr.
sekundarni momenti savijanja i momenti savijanja nastali od ekscentričnih
Pravokutni šuplji profili
spojeva štapova te deformacijska sposobnost spojeva, stoga se u ovom
Rešetkasti nosač radu posebna pozornost posvećuje upravo tim utjecajima. U tom se smislu
Spojevi ilustriraju karakteristični načini modeliranja ponašanja rešetkastih nosača
uz pomoć metode konačnih elemenata (MKE) kako bi se mogle procijeniti
njihove vrijednosti i po potrebi uključiti u proračune. Na numeričkom se
primjeru detaljnije razmatra rešetkasti nosač od šupljih pravokutnih profila
Received (primljeno): 2009-09-01 i sa zavarenim spojevima, te su uspoređene točnosti i dane preporuke za
Accepted (prihvaćeno): 2010-02-19 primjenu pojedinih modela.

considerably in addition to being based on the tradition

1. Introduction of girder design and construction.
In reality, links between individual truss elements are
Truss girders represent a very rational construction usually made stiff (especially welded joints). Deviations
form and facilitate construction making with minimum from centric connecting are common, partly due to
material expenditure and very high degree of efficiency physical inability to create fully centric links, and partly to
regarding construction elements. They are especially facilitate structure construction. In these cases additional
suitable for large-span structures and for greater load stress in truss members occurs (secondary stress), and
transfer in building construction, as well as bridge global truss behaviour is especially influenced by joints
construction. between elements. Because of this, it becomes paramount
Typical truss girder calculation model implies hinge to quantify these phenomena in order to assess their effect
links on joint location, i.e. grid members are influenced on actual construction behaviour. It is known that these
only by axial forces. This approach facilitates calculations phenomena can be tolerated to some extent with regard
126 I. RADIĆ et. al., Design and FEM Modelling... Strojarstvo 52 (2) 125-135 (2010)

Symbols/Oznake
Ai - cross-sectional area of member i (i=0, 1, 2) hi - overall in-plane depth of RHS member i
- površina poprečnog presjeka štapa i (i=0, 1, 2) (i=0, 1, 2)
bi - overall out-of-plane width of RHS member i - ukupna visina pravokutnog kutijastog profila
(i=0, 1, 2) štapa i (i=0, 1, 2) u ravnini rešetkastog nosača
- ukupna širina pravokutnog kutijastog profila ti - wall thickness of RHS member i (i=0, 1, 2)
štapa i (i=0, 1, 2) izvan ravnine rešetkastog - debljina stijenke pravokutnog kutijastog profila
nosača štapa i (i=0, 1, 2)
be,p - effective width for punching shear Ni,Rd - design value of the resistance of the joint,
- djelotvorna širina pri posmičnom proboju expressed with internal axial force in member i
stijenke profila (i=0, 1, 2)
Ce - efficiency parameter, varies depending on the - vrijednost otpornosti spoja izražena preko
type of joints (T, X, K) uzdužne sile u štapu i (i=0, 1, 2)
- koeficijent učinkovitosti, razlikuje se ovisno o β - ratio of diameter or width of brace members to
tipu spoja (T, X, K) that of the chord
e - eccentricity of a joint - omjer promjera ili širina štapova ispune i pojasa
- ekscentricitet u spojevima γ - ratio of the chord width or dameter to twice its
fyi - yield strength of member i (i=0, 1, 2) wall tickness
- granica popuštanja za štap i (i=0, 1, 2) - omjer širine ili promjera pojasa i dvije njegove
debljine stijenke
f (n) - function prestressed chord
- funkcija prednapetosti pojasa η - ratio of the brace member depth to the chord
dameter or width
g - gap between the brace members in a K or N
- omjer visine štapa ispune i širine pojasa
joint, measured along face of chord between the
toes of brace members θi - angle between brace member i and the chord
- razmak između štapova ispune u K ili N (i=0, 1, 2)
spojevima, mjereno između rubova zavara na - kut između štapova ispune i i pojasa (i=0, 1, 2)
štapovima ispune duž lica pojasa

to ductile behaviour and plastic reserves of steel, which in plane or space trusses made of circular, square, or
justifies the application of aforementioned calculation rectangular hollow sections or a combination of open and
model. hollow sections. Characteristic types of truss joints are
However, the articulated joint hypothesis is possible discussed - K, KT, N, T, X, Y, KK, TT, XX, and DY joints.
only if critical parts of truss girder (elements or joints) Calculation rules [2] are based on simplified analytical
have sufficient rotation capacity. Then the secondary models combined with experimental testing, so the
stress originating from bending moment can be neglected regulations are essentially consisted of semi-empirical
in calculations, viz. local joint effects can reduce rotation calculations. In order to describe joint behaviour, it is
capacity of the elements and thus endanger the supposed necessary to note force path, material behaviour, and joint
global behaviour mechanism. Scientific research of stiffness distribution. Mathematically, this is reduced to
plane-loaded truss joint behaviour began in the early basic models of X and K joints and in general terms,
sixties and is based on experimental testing with no brace members force components perpendicular to chord
further theoretical elaboration. Not until lately have the members are discussed, because these force components
existing and new experimental data been associated with cause brace plastification. Resistance of particular joints
theory in greater detail or complexity. [3]. is observable through maximum design resistance of
truss brace members exposed to longitudinal force and/or
bending moment. Thus, it is necessary to check potential
2. Truss girder calculation according to failure locations during calculation process and ascertain
Eurocode regulations possible failure modes of truss girders while considering
local stiffness and joint behaviour. This way, truss chord
In the new European steel construction design optimisation is set as dimensioning goal while controlling
regulations [1] and especially its part [2] this issue is joint stiffness and resistance, since chord members contain
discussed in greater detail, giving basic outlines for up to 3/4 of truss material (with usual truss systems).
design truss girders. Significant innovation is represented Special attention is directed towards compression chord,
by detailed outline of joint calculation in truss girders. It bearing in mind that joint resistance is increased with the
discusses procedures of calculating static joint resistance decrease of chord member local slenderness.
Strojarstvo 52 (2) 125-135 (2010) I. RADIĆ et. al., Design
�������������������������������
and FEM Modelling...����
127

In [2] several possible ways of truss girder joint often complex due to non-linear stiffness distribution
failure are discussed: along the perimeter of the joined brace. When discussing
• chord face failure (by plastification) or plastic failure rectangular hollow sections, difference in stiffness
of the chord cross-section, between section corners and the centre is even greater
than with circular sections, which is why the calculations
• chord side wall failure or chord web failure
are more complex due to different section orientation
by yielding, crushing, or instability under the
possibilities. Consequently, multiple ways of chord face
compression brace member,
failure are possible.
• chord shear failure,
Basic analytic resistance calculation model for this
• punching shear of hollow section chord wall (crack group of joints is based on a yield line model, Figure
initation leading to rapture of the brace member from 1, which assesses plastification of chord members. It
the chord member), should be noted that these formulae are approximate
• brace failure with reduced effective width under and generally give higher stress values regarding chord
crack in the welds or in the brace member, plastification (good estimates are given for mean values
• local buckling failure of the brace member or of a of parameter β) The model is based on equalisation of
hollow section chord member at the joint location. external energy provoked by an external force on the
deformation δ and internal plastification energy together
It should also be noted that in the present version of the with lengths of plastified lines and angles of rotation θ
regulations local slenderness of the cross section is limited (which can be of different configuration, marked in the
more strictly in order to avoid local buckling. Sometimes, figure as “a” and “b”).
multiple criteria are met (for example, chord’s shear
resistance is included in chord plastification formulae, (1)

etc.), so the basic joint failure modes can be reduced to
chord plastification and shear puncture. In order to avoid where:
weld failure it is recommended that the welds be stronger
than joined elements and that the material is not sensitive ,is calculated by unit length, and the meaning
to lamellar tearing. For regulation application purposes, of other designations is visible in Figure 1.
maximum thickness of hollow section wall is limited to
25 mm, while minimum thickness is 2,5 mm. This model is suitable for T, Y, and X joints, while it
can exceptionally be used for K joints. When discussing
Ultimate joint resistance is set by defining either K joints, membrane stress, shear stress, and hardening
maximum load on the force diagram – deformation, or have great influence on their behaviour, so semi-empirical
equivalent load level for preset deformation limit, which approach must be used for calculating this joint type.
is defined according to contemporary works [3] as 3 % of
chord member diameter, i.e. 3 % of chord member width Other important model discusses possibilities of brace
in rectangular and square sections measured at the joint member punching shear. Special attention should be
of brace member and chord member. For serviceability given to the fact that the stiffness on some parts of brace
limit state of use the aforementioned deformation limit member diameter is not even, so it is possible that some
is 1 %. scope parts do not have sufficient deformation capacity
to include the total perimeter of brace member profile.
Formulae for joint resistance are given in terms of Thus, only effective surface is taken into calculations (in
important geometrical parameters which depend on the Y, T, and X joints, brace walls along chord member are
dimensional relationship of brace and chord members: the stiffest parts), and effective width value must be set
• parameter β, as a proportion of average value of brace experimentally.
member diameter/width relative to the corresponding Along these two characteristic failure models,
measuring of chord member problems with the yielding or buckling of chord member
• parameter η, as a proportion of brace member height walls may occur, as well as chord shear failure.
relative to the chord member diameter/width It has been shown that experimental indicators give
• parameter γ, as a proportion of chord diameter/width sufficient information on possible failure modes relative
relative to double thickness of its wall. to parameter β, but the general impression is that there
This paper thoroughly discusses welded truss are many limitations in application of some formulae
constructions made of hollow rectangular cross sections and at the same time many modes of failure. Because of
as common constructional solutions in practice, although this, in order to simplify calculations, a general approach
some parts of the paper are applicable to other types of is adopted where a narrower formulae validity scope is
truss member cross sections. Welded joints are practical targeted in order to reduce the verification of resistance
and thus commonly applied. However, force transfer is to one reliable proof. Also, the approximate graphs are
128 I. RADIĆ et. al., Design and FEM Modelling... Strojarstvo 52 (2) 125-135 (2010)

Figure 1. Characteristic failure modes of truss connections

Slika 1. Karakteristični načini otkazivanja nosivosti spojeva

given to preliminary assess joint effectiveness in the existence of hinge links in truss joints. However, it is
early designing phase so that the behaviour of truss joints advised that additional influences be considered apart
and members is synchronised and that the calculation is from bending moments, that are possible to neglect only
facilitated. In these diagrams, joint resistance is defined in specific cases:
as a fraction of plastic resistance of brace member: • secondary bending moments,
• bending moments due to transverse load between
(2) truss nodes, and
where: • bending moments due to eccentric member
connection in joints.
Ce - is efficiency coefficient that has different
esignations for different joint types (CT,CX,CK), Secondary bending moments are caused by the
rotational stiffness of the joints. They generally depend
θi - is the angle between the brace member and the on absolute and relative member stiffness, static system
chord member, (i.e. conditions of angle alteration between members), and
fyo, to - is the yield strength and wall thickness of chord the magnitude of basic stress. Consequently, secondary
member’s cross section, stress intensity assessment must be associated with the
fyi, ti - is the yield strength and wall thickness of cross applied calculation model of actual construction. For
section of i-chord member, example, Figure 2 shows truss girder segment calculation
f(n) - is the member prestress function (functions model, supposing that the forces in members are of the
as the maximum compression force in the chord same value and that the stiffness and area of the chord
for rectangular hollow sections). are significantly greater than the brace member stiffness.
In the process, rigid link is presumed in the member
In the preliminary truss dimensioning phase the aim intersection, while hinge links are presumed on the
is to ascertain which relationship (fyo to/ fyi ti) should be members’ opposite ends.
foreseen while achieving 100 % joint efficiency in the
process, i.e. that the joint’s bearing capacity does not By using a relatively simple stress and deformation
limit the bearing capacity of the members. Application analysis for this model, an expression can be derived for
example of this type of graph for gapped K-joints is given assessing secondary stress σs in brace members:
in Figure 5.
(3)

3. Bending moment influence on truss

The structure of the abovementioned expression clearly
girders shows that the term (ctg α) represents the contribution
According to [2], it is generally allowed to calculate of a selected static system, σo is basic member stress,
internal forces in girder constructions presuming the while h/l is the member stiffness parameter in the girder
Strojarstvo 52 (2) 125-135 (2010) I. RADIĆ et. al., Design
�������������������������������
and FEM Modelling...����
129

plane (h is the section height in the truss plane while l consideration with joint resistance calculations, and
is the system member length). It should be noted that the total moment is distributed among all the elements
a different expression for secondary stress would have connected in the node. Truss resistance of joints that are
been derived if a different model was used (for example, aditionally loaded with bending moments is generally
a model implying rigid links on the intersection and on resolved the same way as joints with axial stress, bearing
the free ends of members). in mind that the chord plastification and punching shear
mechanisms are somewhat modified.

4. Numeric modelling of truss girders

Past analyses make it apparent that the truss girder
behaviour modelling should consider global, as well
as local behaviour of construction elements and joints.
The Finite Element Method (FEM), depending on
the requested accuracy and correctness of the model,
differentiates betweeen:
• member model of girder joints with or without
considering possible secondary influences, or
influences due to joint eccentricity,
Figure 2. Example of design model for estimation of • space girder models which can be completely made
secondary stress of shell type space elements, or a combination of
Slika 2. Primjer proračunskog modela za procjenu shell and beam elements,
sekundarnih naprezanja • isolated truss girder joint models.
Beam model application is the simplest and the
Expression 3 shows that secondary stress will be lower most acceptable in engineering terms, and it is generally
with members whose height in the truss plane h is smaller possible to conduct linear or non-linear analysis (material
than the member length l, so this ratio is very important and/or geometrical). The most widely applied model in
when assessing the magnitude of secondary stress due to practice is the one implying hinge links in nodes, when the
rigid links. Thus, in [2] is also stated that the observed result is only axial forces in members. Then the potential
secondary stress can be neglected in the calculations if bending stress resulting from more precise analyses is
l/h ratio has corresponding values (for example, for truss considered secondary, since it does not influence the
constructions in building construction the minimum ratio balance and is generally possible to be neglected (if it
value for the stress to be neglected is 6). It should be occurs within reasonable limits). However, if the truss
noted that in case of fatigue danger this stress can have is made as a frame system (node link is “rigid” instead
significant influence and must be taken into account. of hinge), then the bending stress calculated in this way
Moments originating from eccentric joint links can is not seen as secondary stress, since it can influence
be neglected when calculating the resistance of chord the longitudinal force values in such models. In the
and brace tension resistance, while in joint resistance case when the joint eccentricities are modelled, one of
calculations they can be neglected only if the eccentricity the models in Figure 3 is used. By selecting different
element e is in the following interval: links between brace members or between brace and
chord members, link eccentricity bending moment can
(4)
be influenced (in accordance with previously described
moment reallocation [2]). With this type of modelling,
where ho is the height of chord member in the truss
it is not possible to consider local effects that appear in
plane.
joints as consequence of their deformation characteristics,
However, with compression chords the link eccentricity not even by applying non-linear (elastoplastic) analysis.
must always be taken into account. For eccentricities that In other words, if material elastoplasticity was taken into
fall under abovementioned parameters, they are taken consideration, it would consider only stress redistribution
only for compression chord resistance calculations in (i.e. gradual bending moment decrease in chord elements
such a way that the total moment is distributed between due to rotational stiffness with simultaneous increase of
chord members on each side of the joint proportional longitudinal force), caused by ductile behaviour of steel.
to their relative stiffness I/l (where l is system length of
Truss space models in which the truss elements
the member). If the eccentricity value e exceedes values
are modelled by shell-type plane finite elements are
given in the interval above, then it has to be taken into
very precise and they include global and local system
130 I. RADIĆ et. al., Design and FEM Modelling... Strojarstvo 52 (2) 125-135 (2010)

behaviour. These models are used only for research so it is not necessary to consider this mode of failure. In
purposes because of creation complexity of this model, the process of overlap joints and gap joints construction,
which largely depends on truss cross section types, on gap and eccentricity tollerance guidelines given in [2]
the static system complexity (i.e. space truss), and the were followed. Software packages used for modelling
calculation duration (especially non-linear analysis). are Autodesk Robot [5] for beam model construction,
An acceptable modelling approach could be the and ADINA [6] for space model construction (shell-type
combination of beam and plane elements in such a way finite elements with 4 nodes were used).
that the joint area is modelled by plane elements, and other Characteristic truss details are given in Table 1.
areas by beam elements. The connection is made through Figure 5 shows the procedure of approximate
rigid links (i.e. master-slave). This way the model creation definition of critical joint efficiency (joint 4 in Figure 2)
complexity is significantly reduced, calculation results by using preliminary diagrams. Table 1 data show that the
include both global and local joint element behaviour, critical joint in the truss girder is joint 2 in Figure 4, since
and most importantly, modelled joints are exposed to the calculated load (according to linear theory) achieves
actual boundary conditions existing in the construction its 100-percent efficiency, while other joints under same
itself. Aforementioned advantage is especially related load are less used. The relevant failure mode in the
to so-called isolated models that include only specific process was chord member face plastification. Force in
(usually critical) joints, but only when it is impossible to compression brace member 1 for achieving 100-percent
fully include actual influences that appear in that joint at efficiency in joint 2 was N1,Rd=153.7 kN. Figure 5 shows
its actual location in the construction [4]. that by applying approximate procedure, with the aid
of finished diagrams, a more conservative force for this
compression diagonal can be derived (N1,Rd=137.5 kN).
5. Numeric example
Furthermore, by using FEM, the following calculations
Truss girder modelling types using FEM shall be have been derived:
explained and exemplified by means of a simple truss
construction. Calculation results shall be critically • non-linear truss calculations in beam models together
analysed and compared in relation to the codified with:
approach [2]. - presumed rigid centric links between members,
Truss sketch together with measures, used cross - presumed rigid overlap joints,
sections, and load distribution is shown in Figure 4. Initial - presumed rigid joints with gap.
force is 200 kN, and the presented calculation results • non-linear calculations on a 3D truss model made by
show the factor which is multiplied by initial force in using shell-type finite elements,
order to assert actual load. Steel quality is S235, and the • non-linear calculations on the combined model
models presume idealised elastoplastic steel behaviour consisting of beam elements and shell elements,
diagram. Welding material exhibits the same properties • non-linear calculations of isolated truss girder
as the basic material and sufficient weld size is presumed, joints.

Figure 3. Various modelling methods of truss bar models

Slika 3. Varijante modeliranja rešetki štapnim modelima
Strojarstvo 52 (2) 125-135 (2010) I. RADIĆ et. al., Design
�������������������������������
and FEM Modelling...����
131

Figure 4. Schematic representation of truss girder used in analysis

Slika 4. Shematski prikaz rešetkaste konstrukcije korištene u proračunima

Figure 6 shows force factor-displacement diagrams,

where there is a global vertical displacement of joint 3 (z
axis). It is visible that the results for various calculation
models are very similar, which is expected considering
that the eccentricity and secondary stress limitation
guidelines were followed.
However, it is apparent that there is still a significant
difference in the limit load of space and beam models due
to possible inclusion of local joint behaviour in the space
models (Figure 7). Consequently, according to 3D space
models, the force in critical diagonal 1 is N1,Rd=229,8 kN,
while for beam models it is N1,Rd=289,6 kN (but that force
is actually related to global failure of diagonal 1 and not
its joints). By comparing obtained results and the codified
procedure it is visible that the formulae in [2] give
somewhat conservative estimates of joints resistance (the
Figure 5. Diagram of efficiency of braces for K joints with difference is around 33 % compared to 3D space truss
gap models), which is expected considering predescribed
Slika 5. Dijagrami za određivanje učinkovitosti K spojeva s
suppositions for their defining. Similar observations can
razmakom
be found in other works, for example [7].

Displacement / Pomak uz, cm Displacement / Pomak uz, cm

a) b)
Figure 6. Force factor – displacement diagram for analyzed models
Slika 6. Dijagram faktor sile - pomak za korištene modele
132 I. RADIĆ et. al., Design and FEM Modelling... Strojarstvo 52 (2) 125-135 (2010)

It is also visible that the results for the combined 3D purposes all 3D combined model joints are modelled by
model and the full 3D shell are almost identical (Figures shell elements, but that does not have to be the case.
6a and 8), which justifies this type of modelling example It is sufficient to model critical details that are easy to
since it is much simpler. It should be noted that for exaple determine preliminarily (as described earlier).

Table 1. Codified design of characteristic truss connections according [2]

Tablica 1. Kodificirani proračun otpornosti karakterističnih detalja prema [2]

EFFICIENCY OF K - JOINTS / UČINKOVITOST K SPOJEVA

PLASTIC FAILURE OF THE CHORD FACE / PLASTIFIKACIJA LICA POJASA

fy0 = 235 N/mm2

JOINT 2 / SPOJ 2
fy0 = fy1 = 235 N/mm2 kn = 1,035 → kn = 1,00
b0 =h0 = 100 m γ = 10,00
t0 = 5,0 mm β = 0,60
b1 = h1 = 70 mm
t1 = 5,0 mm N1,Rd = 152,45 kN
b2 = h2 = 70 mm
t2 = 3,2 mm 100%
θ = 40,60°
JOINT 3 / SPOJ 3
fy0 = fy1 = 235 N/mm2 kn = 0,964
b0 =h0 = 100 m γ = 10,00
t0 = 5,0 mm β = 0,60
b1 = h1 = 70 mm
t1 = 5,0 mm N1,Rd = 144,22 kN
b2 = h2 = 70 mm
t2 = 3,2 mm 53,3%
θ = 40,60°
JOINT 4 / SPOJ 4
fy0 = fy1 = 235 N/mm2 kn = 1,072 → kn = 1,00
b0 =h0 = 100 m γ = 10,00
t0 = 5,0 mm β = 0,70
b1 = h1 = 70 mm
t1 = 5,0 mm N1,Rd = 177,86 kN

43,2%
θ = 40,60°
Strojarstvo 52 (2) 125-135 (2010) I. RADIĆ et. al., Design
�������������������������������
and FEM Modelling...����
133

Figure 9 shows details of joint 2 isolated model, A relatively good diagram overlap can be noticed, and
and Figure 10 shows parallel results of 3D shell force- once more the emphasis should be put on the importance
displacement results, the combined model results, and of the proper boundary condition modelling in isolated
the isolated model results. The designation uz,r marks node models.
the greatest relative shift of the chord member wall with
welded brace members relative to the opposite wall.

Figure 7. Part of 3D model (connection 3)

Slika 7. Prikaz dijela 3D modela (spoj 3)

Figure 8. View of the combined model

Slika 8. Prikaz kombiniranog modela
134 I. RADIĆ et. al., Design and FEM Modelling... Strojarstvo 52 (2) 125-135 (2010)

Figure 9. View of the isolated model of connection 1

Slika 9. Prikaz izdvojenog modeliranja detalja 1

this construction type using the Finite Element Method

(FEM). All modelling possibilities have been analysed on
a numeric example of a simpler truss and the calculation
results have been compared. Bilinear steel behaviour
has been preset and non-linear calculations have been
conducted.
Calculation results point to possible areas of
individual model application, as well as common need
to use combined models considering the global and local
construction behaviour. It especially concerns cases
not falling under strict tollerancies and dimensional
limitations of Eurocode regulations. This system, after
using FEM, exhibits slightly greater limit stress values
than is bound by regulations.
Based on the aforementioned findings, the final
Figure 10. Diagram force – relative displacement for various conclusion is that, considering a relatively narrow area of
models regulation validity [2] and its conservativeness, this area
Slika 10. Dijagram sila-relativni pomak za različite modele requires further experimental and numeric tests.

6. Conclusion
references
The paper gives a survey of truss girder calculation
model according to new Eurocode regulations [1, [1] European Committee for Standardization (CEN):
2]. Joint design rules in [2] are set in semi-empirical EN 1993-1-1, Eurocode 3: Design of steel structures
formulae and are valid in very limited dimension and – Part 1-1: General rules and rules for buildings,
geometry conditions of truss girders. Therefore the 2005.
need of more detailed calculations arises frequently in [2] European Committee for Standardization (CEN):
engineering practice, especially on local joint level. Thus, EN 1993-1-8, Eurocode 3: Design of steel structures
this paper gives and comments on modelling modes of – Part 1-8: Design of joints, 2005.
 Strojarstvo 52 (2) 125-135 (2010) I. RADIĆ et. al., Design
�������������������������������
and FEM Modelling...����
135

[3] WARDENIER, J.: Hollow sections in structural [6] ADINA user interface command reference manual,
applications, Comitè International pour le Volume I: ADINA Solids & structures model
Dèveloppement et l’Etude de la Construction definition, Report ARD 09-2, May 2009., ADINA
Tubulaire, CIDECT; 2000 R&D Inc.
[4] DEL COZ DÍAZ, J.J.; GARCÍA NIETO, P.J.; [7] CHRISTITSAS, A.D.; PACHOUMIS, D.T.;
FERNÁNDEZ RICO, M.; SUÁREZ SIERRA, KALFAS, C.N.; GALOUSSIS, E.G.: Fem analysis
J.L.: Non-linear analysis of the tubular ‘heart’ joint of cenventional and square bird-beak SHS joint
by FEM and experimental validation, Journal of subject to in-plane bending moment-experimental
Constructional Steel Research, Volume 63, Issue 8, study, Journal of Constructional Steel Research 63
August 2007, 1077-1090 (2007), 1361-1372
[5] AUTODESK ROBOT STRUCTURALANALYSIS:
Getting started guide,2009., Autodesk, Inc.

View publication stats